Identification of a fluorescent small-molecule enhancer for therapeutic autophagy in colorectal cancer by targeting mitochondrial protein translocase TIM44

Article abstract of DOI:10.1136/gutjnl-2016-311909

Objective As the modulation of autophagic processes can be therapeutically beneficial to cancer treatment, the identification of novel autophagic enhancers is highly anticipated. However, current autophagy-inducing anticancer agents exert undesired side effects owing to their non-specific biodistribution in off-target tissues. This study aims to develop a multifunctional agent to integrate cancer targeting, imaging and therapy and to investigate its mechanism. Design A series of mitochondria-targeting near-infrared (NIR) fluorophores were synthesised, screened and identified for their autophagy-enhancing activity. The optical properties and biological effects were tested both in vitro and in vivo. The underlying mechanism was investigated using inhibitors, small interfering RNA (siRNA), RNA sequencing, mass spectrometry and human samples. Results We have screened and identified a new NIR autophagy-enhancer, IR-58, which exhibits significant tumour-selective killing effects. IR-58 preferentially accumulates in the mitochondria of colorectal cancer (CRC) cells and xenografts, a process that is glycolysis-dependent and organic anion transporter polypeptide-dependent. IR-58 kills tumour cells and induces apoptosis via inducing excessive autophagy, which is mediated through the reactive oxygen species (ROS)-Akt-mammalian target of rapamycin (mTOR) pathway. RNA sequencing, mass spectrometry and siRNA interference studies demonstrate that translocase of inner mitochondrial membrane 44 (TIM44)-superoxide dismutase 2 (SOD2) pathway inhibition is responsible for the excessive ROS, autophagy and apoptosis induced by IR-58. TIM44 expression correlates positively with CRC development and poor prognosis in patients. Conclusions A novel NIR small-molecule autophagy-enhancer, IR-58, with mitochondria-targeted imaging and therapy capabilities was developed for CRC treatment. Additionally, TIM44 was identified for the first time as a potential oncogene, which plays an important role in autophagy through the TIM44-SOD2-ROS-mTOR pathway.

Full text of DOI:10.1136/gutjnl-2016-311909

Gut Online First, published on November 14, 2016 as 10.1136/gutjnl-2016-311909
GI cancer
ORIGINAL ARTICLE
Identification of a fluorescent small-molecule
enhancer for therapeutic autophagy in colorectal
cancer by targeting mitochondrial protein
translocase TIM44
Yinghui Huang,1 Jie Zhou,2 Shenglin Luo,1 Yang Wang,1 Jintao He,1 Peng Luo,1
Zelin Chen,1 Tao Liu,1 Xu Tan,1 Juanjuan Ou,2 Hongming Miao,2 Houjie Liang,2
Chunmeng Shi1
▸ Additional material is
ABSTRACT
published online only. To view
please visit the journal online
(http://dx.doi.org/10.1136/
gutjnl-2016-311909).
Objective As the modulation of autophagic processes
can be therapeutically beneficial to cancer treatment, the
identification of novel autophagic enhancers is highly
anticipated. However, current autophagy-inducing
anticancer agents exert undesired side effects owing to
their non-specific biodistribution in off-target tissues.
This study aims to develop a multifunctional agent to
integrate cancer targeting, imaging and therapy and to
investigate its mechanism.
Significance of this study
1Institute of Combined Injury,
State Key Laboratory of
Trauma, Burns and Combined
Injury, Chongqing Engineering
Research Center for
What is already known on this subject?
▸ The modulation of autophagic processes can be
therapeutically beneficial for enhancing
tumouricidal effects during cancer treatment,
but current autophagy-inducing agents for
cancer therapy suffer from a lack of tumour
selectivity and exert undesired side effects.
▸ To develop tumour-targeted therapeutic agents,
current strategies primarily explore the chemical
conjugation of various functional agents such as
tumour-specific ligands, contrast agents and
anticancer drugs. Structure-inherent cancer
targeting small-molecule agents without the need
for chemical conjugation is highly anticipated.
▸ Translocase of inner mitochondrial membrane
44 (TIM44) has been recently identified as an
important player in the regulation of reactive
oxygen species (ROS) and mitochondrial uptake
of the nuclear-encoded protein, but the role
and mechanism of TIM44 in autophagy and
colorectal cancer (CRC) progression remain
unknown.
Nanomedicine, College of
Preventive Medicine, Third
Military Medical University,
Chongqing, China
Design A series of mitochondria-targeting near-infrared
(NIR) fluorophores were synthesised, screened and
identified for their autophagy-enhancing activity. The
optical properties and biological effects were tested both in
vitro and in vivo. The underlying mechanism was
investigated using inhibitors, small interfering RNA (siRNA),
RNA sequencing, mass spectrometry and human samples.
Results We have screened and identified a new NIR
autophagy-enhancer, IR-58, which exhibits significant
tumour-selective killing effects. IR-58 preferentially
accumulates in the mitochondria of colorectal cancer (CRC)
cells and xenografts, a process that is glycolysis-dependent
and organic anion transporter polypeptide-dependent. IR-
58 kills tumour cells and induces apoptosis via inducing
excessive autophagy, which is mediated through the
reactive oxygen species (ROS)-Akt-mammalian target of
rapamycin (mTOR) pathway. RNA sequencing, mass
spectrometry and siRNA interference studies demonstrate
that translocase of inner mitochondrial membrane 44
(TIM44)-superoxide dismutase 2 (SOD2) pathway inhibition
is responsible for the excessive ROS, autophagy and
apoptosis induced by IR-58. TIM44 expression correlates
positively with CRC development and poor prognosis in
patients.
2Department of Oncology,
Southwest Cancer Center,
Southwest Hospital, Third
Military Medical University,
Chongqing, China
Correspondence to
Professor Chunmeng Shi,
Institute of Combined Injury,
State Key Laboratory of
Trauma, Burns and Combined
Injury, Chongqing Engineering
Research Center for
Nanomedicine, College of
Preventive Medicine, Third
Military Medical University,
Chongqing 400038, China;
shicm@sina.com
What are the new findings?
▸ A novel tumouricidal autophagy-inducing
small-molecule dye, IR-58, is identified as
preferentially accumulating in the mitochondria
of tumour cells and specifically targets tumour
sites in vivo.
Received 22 March 2016
Revised 6 October 2016
Accepted 18 October 2016
▸ IR-58 inhibits cancer growth and induces
apoptosis via inducing excessive autophagy in
vitro and in vivo, which is mediated through
the ROS-Akt-mammalian target of rapamycin
(mTOR) pathway.
▸ TIM44 is characterised for the first time as
playing an important role in suppressing
autophagy through the TIM44-superoxide
dismutase 2 (SOD2)-ROS-mTOR pathway, which
is responsible for IR-58-induced excessive ROS
and apoptosis.
Conclusions A novel NIR small-molecule autophagy-
enhancer, IR-58, with mitochondria-targeted imaging and
therapy capabilities was developed for CRC treatment.
Additionally, TIM44 was identified for the first time as a
potential oncogene, which plays an important role in
autophagy through the TIM44-SOD2-ROS-mTOR pathway.
To cite: Huang Y, Zhou J,
Luo S, et al. Gut Published
Online First: [please include
Day Month Year]
doi:10.1136/gutjnl-2016-
311909
▸ TIM44 expression correlates positively with CRC
development and prognosis, suggesting that
TIM44 is a potential therapeutic target in
human CRC.
INTRODUCTION
Colorectal cancer (CRC) occurrence and mortality
are increasing, while successful tumour therapy
remains elusive for many patients.1 Recent studies
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
1
Copyright Article author (or their employer) 2016. Produced by BMJ Publishing Group Ltd (& BSG) under licence.

GI cancer
reported methods, the key intermediate (compound 5, yield
60%) was successfully synthesised using an amidation reaction
between compound 2 and 4 in the presence of Et3N for the
base catalysis. The chemical structure of compound 5 was char-
acterised using 1H-nuclear magnetic resonance (1H-NMR).
Subsequently, compound 6, in the form of an indole quaternary
ammonium salt, was prepared using an N-alkylation reaction
Significance of this study
How might it impact on clinical practice in the
foreseeable future?
▸ IR-58 may represent a novel autophagy-inducing
tumouricidal small-molecule enhancer and the
TIM44-SOD2-ROS-mTOR pathway may be a potential target
for clinical CRC treatment. This work may also provide a
potential strategy to develop multifunctional small-molecule
agents for imaging-guided CRC treatment, which can
enhance the specificity of tumour targeting and decrease
potential side effects in off-target tissues, along with
enabling real-time monitoring of the tumour response to
therapy.
between 2,3,3-trimethyl-3H indole and compound
5 in
1,2-dichlorobenzene. Finally, compound 6 reacted with com-
pound 7 in a condensation reaction, with sodium acetate as the
base catalyst to synthesise IR-58 (1.63 g, yield 30%, green sticky
solid). The structural characterisation was determined using
1H-NMR and high resolution mass spectrometry (HRMS).
The absorption and fluorescence spectra of IR-58 were inves-
tigated in methanol (MeOH) and 10% foetal bovine serum
(FBS; figure 1C, D) and compared with those of indocyanine
green (ICG), which is a clinically available heptamethine dye.24
The absorption and emission peaks of IR-58 as well as those of
ICG were in the NIR region (700–900 nm). Notably, the fluor-
escence intensity of IR-58 in 10% FBS was higher than that of
have identified that abnormalities in autophagy are closely
linked to human diseases including CRC and autophagy is con-
sidered a potential target for cancer therapy.2–5 Several clinically
approved and experimental antitumour agents have been shown
to induce autophagy-mediated cell death in various types of
cancer and drugs augmenting autophagy hold promise for
ICG. IR-58 also showed
a good stability in 10% FBS
(figure 1E), suggesting that it has potential as a new NIR probe
in biomedical imaging.
IR-58 preferentially accumulates in the mitochondria of
tumour cells in a glycolysis-dependent and organic anion
transporter polypeptide-dependent manner
cancer
treatment.6–11
However,
most
current
autophagy-inducing agents exert undesired side effects owing to
non-specific biodistribution in off-target tissues, highlighting an
urgent need to develop highly tumour-specific and efficient
inducers of autophagy.5 12 13
To obtain tumour-targeted therapeutic agents, current strat-
egies mainly explore the chemical conjugation of various func-
tional agents such as tumour-specific ligands, contrast agents
and anticancer drugs.14–19 However, chemical conjugation is
The potential of IR-58 for preferential tumour accumulation
was investigated in HT-29 and HCT116 CRC xenografts (figure
2A, B). Nude mice with subcutaneous human HT-29 and
HCT116 tumour xenografts were established and imaged after
a single intravenous dose of IR-58 at 0.2 mg/kg. Figure 2A, B
show that intense fluorescence signals associated with the
implanted tumours can be clearly visualised after IR-58 injec-
tion, with low background interfering fluorescence. The tumour
cytosolic accumulation of IR-58 iodide was confirmed using
histopathological analysis of the tumour specimens by confocal
microscopy (figure 2C). Similar results were also shown in
Sprague-Dawley rats (see online supplementary figure S1).
A subcellular localisation assay, using colocalisation with a
mitochondrial tracker (MitoTracker Green, Molecular Probes,
Eugene, Oregon, USA) in HT-29 and HCT116 cells, shows that
IR-58 exclusively accumulates in the mitochondria of tumour
cells (figure 2D). As the mitochondrion is the energy centre of
the cell and IR-58 locates to the mitochondria, we speculated
that the uptake of IR-58 is energy-dependent. Thus, HT-29 and
HCT116 cells were exposed to a low temperature (0°C) for
30 min prior to IR-58 treatment, and IR-58 uptake significantly
decreased (figure 2E, F). Because cellular metabolic energy is
mainly supplied by glycolysis and oxidative phosphorylation,
usually limited by low efficiency of drug delivery, difficulties in
large-scale preparation and potential toxicity.14
To over-
20 21
come these limitations, exploring novel small molecules with
intrinsic structure-inherent targeting properties would be an
ideal strategy and has attracted increasing attention.14 Very
recently, we have identified a couple of near-infrared (NIR) hep-
tamethine dyes with preferential accumulation in the mitochon-
dria of a broad spectrum of cancer cells, without chemical
conjugation of target ligands. The molecules exhibit unique
optical properties to achieve a significant signal-to-noise ratio
and hold promise as potential agents for tumour-targeted
imaging.14 Based on the previous work, in this study we have
synthesised a series of new derivatives modified with various
N-alkyl side chains on a mitochondria-targeted heptamethine
core and successfully identified a novel NIR dye, IR-58, which
exhibits significant tumour killing effects through inducing
HT-29
and
HCT116
cells
were
pretreated
with
excessive
autophagy.
IR-58
represents
a
novel
2-deoxy-d-glucose (2-DG; an inhibitor of glycolysis) and oligo-
mycin (an inhibitor of oxidative phosphorylation) for 30 min,
respectively. Figure 2E, F show that 2-DG significantly sup-
presses the uptake of IR-58, indicating that uptake of IR-58 is
glycolysis-dependent. Oligomycin did not have this effect.
Moreover, we also identified that organic anion transporter
polypeptide (OATP) mediates the uptake of IR-58 by tumour
cells. When HT-29 and HCT116 cells were pretreated with sul-
fobromophthalein (BSP, a competitive inhibitor of OATP trans-
porters) for 30 min, the fluorescence values were significantly
lower than those observed for untreated cells, which suggests
that the uptake of IR-58 is OATP-dependent (figure 2G, H). In
addition, endocytosis is one of the most common methods of
autophagy-enhancer candidate for potential cancer-targeted
imaging and therapy.
RESULTS
Chemical synthesis, screening and optical properties of
IR-58
We have synthesised a series of heptamethine cyanine dyes by
modification of N-alkyl side chains based on a mitochondria-
targeted heptamethine core. Of these, IR-58 was discovered to
be capable of inducing autophagy in CRC cells (figure 1A). The
chemical synthesis procedure for IR-58 is shown in figure 1B.
After synthesis of compounds 222 and 423 using previously
2
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
Figure 1 Chemical synthesis, analysis of autophagy-inducing effects and optical properties of IR-58. (A) HT-29 and HCT116 cell lines were treated
with a series of heptamethine cyanine dyes for 24 hours, and harvested for detection of autophagy markers (LC3 and SQSTM1) using western blot
analysis. β-Actin was used as the loading control. (B) Synthetic route and chemical structure of IR-58. (C) The absorption spectra of 2 μM IR-58 and
indocyanine green (ICG) in 10% foetal bovine serum (FBS) and methanol. (D) Fluorescence image of IR-58 and ICG in 10% FBS and methanol.
(E) Fluorescence stability of IR-58 in 10% FBS at 37°C for 48 hours. Each valid point and vertical bar represents mean SD.
cell uptake for drugs. HT-29 and HCT116 cells were pretreated
with chlorpromazine or carbonyl cyanide m-chlorophenyl
hydrazone (CCCP; both of which are classic endocytosis inhibi-
tors), but the fluorescence values demonstrated no obvious
changes, suggesting that the uptake of IR-58 is
endocytosis-independent (figure 2G, H).
time-dependently in HT-29 and HCT116 cells (figure 4A,B;
p<0.05). Autophagy was also induced in A549 and HepG2 cells
(see online supplementary figure S4). Downregulation of the
sequestosome 1 (SQSTM1) autophagy substrate in IR-58-treated
cells (figure 4A, B) suggests that autophagy is induced by
IR-58.25 TEM images show autophagic vacuole formation in
IR-58-treated cells, suggesting that autophagy is induced by
IR-58 (figure 4C). GFP-LC3 puncta were observed in
IR-58-treated GFP-LC3-expressing HT-29 and HCT116 cells
using confocal microscopy and their presence was significantly
inhibited by 3-methyladenine (3MA; an autophagy inhibitor;
figure 4D). IR-58-induced LC3II upregulation was also reversed
by 3MA (figure 4E; p<0.05). In addition, the IR-58-induced
inhibition of HT-29 and HCT116 cell proliferation (figure 4F)
and induction of apoptosis (figure 4G, H) were reversed by 3MA
(p<0.05). Small interfering RNA (siRNA) against Atg7 (an
important autophagy-associated gene) was used to inhibit Atg7
expression (figure 4I), and IR-58-mediated suppression of cell
proliferation was partly reversed by siAtg7 (figure 4J; p<0.05).
We investigated the effect of IR-58 on non-cancerous cells in
vitro. The uptake of IR-58 in non-tumourigenic CCD841CoN,
HBE and L02 cells was much lower than in HT-29, HCT116,
A549 and HepG2 tumourigenic cells (p<0.01; see online
supplementary figure S5A), indicating that the accumulation of
IR-58 in non-cancerous cells was lower than in cancer cells. In
addition, the IC50 of CCD841CoN (9.515 μM), HBE
(9.661 μM) and L02 (7.680 μM) cells exposed to IR-58 was
much higher than that of the HT-29 (2.888 μM), HCT116
(4.084 μM), A549 (1.739 μM) and HepG2 (3.281 μM) cells
(figure 3A and see online supplementary figure S5B). Upon
IR-58 kills tumour cells and induces cell apoptosis
In addition to the optical properties and the subcellular uptake
of IR-58, we investigated its biological effects on tumour cells.
Figure 3A, B show that IR-58 can inhibit HT-29 and HCT116
cell proliferation dose-dependently (p<0.001) and time-
dependently (p<0.001). However, cell cycle arrest was not sig-
nificantly influenced by IR-58 at 24 and 48 hours (p>0.05; see
online supplementary figure S2), suggesting that cell death may
be responsible for the decrease in cell numbers. As shown in
figure 3C, D, IR-58 induced apoptosis dose-dependently
(p<0.001). The apoptosis was also visualised using transmission
electron microscopy (TEM; figure 3E). Moreover, apoptosis
markers such as poly ADP-ribose polymerase (PARP), cleaved
caspase 3 (c-caspase 3) and cleaved caspase 9 (c-caspase 9) were
all induced dose-dependently by IR-58 (figure 3F).
Furthermore, we also found that IR-58 could kill lung cancer
cells (A549) and liver cancer cells (HepG2) and induce apop-
tosis in these cell types (see online supplementary figure S3).
IR-58 kills tumour cells via inducing excessive autophagy in
vitro
IR-58-induced expression of the microtubule-associated protein
1 light chain 3 (LC3)II autophagy marker dose-dependently and
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
3

GI cancer
Figure 2 IR-58 specifically targets tumour tissue and locates to cancer cell mitochondria, in a process which is dependent on glycolysis and
organic anion transporter polypeptide (OATP). (A and B) Nude mice with human HT-29 and HCT116 subcutaneous tumour xenografts were
subjected respectively to near-infrared (NIR) imaging and X-ray imaging, after a single-dose intravenous administration of IR-58 at 0.2 mg/kg.
Organs and xenografts were imaged using a NIR imaging system. (C) Xenograft sections shown in (B) were incubated with 40,6-diamidino-2-
phenylindole (DAPI), for 5 min, and washed with phosphate buffered saline prior to laser confocal scanning microscope imaging. (D) Colocalisation
of IR-58 with a mitochondria-specific tracker (MitoTracker Green) in HT-29 and HCT116 cells, imaged using a confocal microscope. Nuclei were
stained with Hoechst 33258. (E) HT-29 and HCT116 cells were pretreated with 0.1% dimethyl sulfoxide, 2-deoxy-D-glucose (2-DG) (150 mM),
oligomycin (1 μM) or ice for 30 min, and then treated with 10 μM IR-58 for 15 min prior to confocal microscopy observations. (F) Fluorescence
intensities of the samples presented in (E) were obtained from three independent measurements. (G) HT-29 and HCT116 cells were pretreated with
vehicle control, sulfobromophthalein (BSP) (250 μM), chlorpromazine (20 μM) or carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (20 μM) for
30 min, then treated with IR-58 for 15 min prior to confocal microscopy observations. (H) Fluorescence intensities of the samples presented in (G)
were obtained from three independent measurements. **p<0.01, ***p<0.001.
exposure to IR-58, the apoptosis rate (figure 3C, D, see online
supplementary figure S5C,D) and autophagy marker protein
levels (p62 and LC3; figure 4A and see online supplementary
figure S5E) of CCD841CoN, HBE and L02 cells changed less
than those of HT-29, HCT116, A549 and HepG2 cells
(p<0.05). These results support IR-58 as a cancer targeting
agent with lower specificity and side effects in non-cancerous
cells.
4
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
Figure 3 IR-58 kills tumour cells and induces apoptosis. (A) HT-29 and HCT116 cells were treated with various doses of IR-58 for 48 hours and
cell viability was measured using CCK-8 kit. (B) HT-29 and HCT116 cells were treated with 10 μM IR-58 at different times and OD450 was measured
using CCK-8 kit. (C) HT-29 and HCT116 cells were treated using various doses of IR-58 for 24 hours and harvested for detection of apoptosis using
flow cytometry. (D) One-way analysis of variance statistics of the cytometry data derived from samples presented in (C), using three independent
measurements. (E) HT-29 and HCT116 cells were treated with 10 μM IR-58 for 24 hours and observed for apoptosis using transmission electron
microscopy. (F) HT-29 and HCT116 cells were treated with various doses of IR-58 and apoptosis markers were detected using western blot analysis.
β-Actin was taken as the loading control. *p<0.05, **p<0.01, ***p<0.001. PARP, poly ADP-ribose polymerase.
the nude mice in different treatment groups (see online
supplementary figure S6B).
IR-58 kills tumour cells via inducing excessive autophagy in
vivo
IR-58 significantly suppressed the proliferation of HT-29 cells in
nude mice, which was partly reversed by 3MA (figure 5A, B;
p<0.001). Xenograft weight was inhibited by IR-58, and this
was reversed by 3MA (figure 5C). However, the body weight of
the nude mice demonstrated no significant differences among
the different groups (figure 5D), implying a desirable safety in
IR-58-treated mice. H&E staining shows that the organs of the
mice (lung, liver, spleen, heart, kidney, colon and muscle) were
not significantly influenced by IR-58 (see online supplementary
figure S6A). TEM imaging shows that autophagy and apoptosis
were induced in the IR-58-treated mice (figure 5E). As shown in
figure 5F, IR-58 induced the expression of autophagy markers
in vivo, and this was inhibited by 3MA. Figure 5G demonstrates
that IR-58 could induce apoptosis, which was repressed by
3MA, suggesting that inhibition of autophagy could suppress
apoptosis induced by IR-58. These results were validated using
immunohistochemical detection (figure 5H). As expected, LC3II
levels demonstrated no significant change among the organs of
IR-58 induces autophagy and cell death through the reactive
oxygen species-Akt-mammalian target of rapamycin
pathway
As mitochondria are the main source of reactive oxygen species
(ROS) and IR-58 locates to the mitochondria, we investigated
ROS production. ROS was significantly enhanced by IR-58 treat-
ment in a dose-dependent manner in HT-29 and HCT116 cells
(figure 6A, B). N-Acetylcysteine (NAC), a ROS scavenger, could
significantly reverse the IR-58-induced increase in ROS (figure 6C,
D). As expected, the proliferation of CRC cells was suppressed by
IR-58, and this was reversed by NAC (figure 6E). These results
demonstrate that ROS is required in IR-58-mediated cell death.
NAC rescued the increased LC3II expression induced by IR-58
(figure 6F; p<0.05), which implies that ROS is the initiating factor
of IR-58-mediated autophagy. As the Akt-mammalian target of
rapamycin (mTOR) pathway is the downstream effector of ROS
and an initiator of autophagy, we carried out a western blot
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
5

GI cancer
Figure 4 IR-58 kills tumour cells through inducing excessive autophagy in vitro. (A) HT-29 and HCT116 cells were treated with various doses of
IR-58 for 24 hours, and total proteins were harvested for detection of autophagy markers (LC3, SQSTM1 and LAMP1). (B) HT-29 and HCT116 cells
were treated with 10 μM IR-58 for different times and total proteins were harvested for detection of autophagy markers (LC3 and SQSTM1). (C)
HT-29 and HCT116 cells were treated with 10 μM IR-58 and observed for autophagy using transmission electron microscopy. AV: autophagy
vacuoles. (D) HT-29 and HCT116 cells were transfected with a GFP-LC3 plasmid, and then treated with vehicle control, 4 mM 3-methyladenine
(3MA), 10 μM IR-58 with or without 3MA for 24 hours. The formation of GFP-LC3 puncta was observed using a laser confocal scanning microscope.
(E) HT-29 and HCT116 cells were treated with vehicle control, 3MA (4 mM), IR-58 (10 μM) with or without 3MA for 24 hours and LC3 was detected
using western blot analysis. (F) HT-29 and HCT116 cells were treated as in (E) for 48 hours, and cell viability was measured using CCK-8 kit. (G)
Cells were treated as in (E) and apoptosis was detected using flow cytometry. (H) A statistical analysis of the data derived from (G), using three
independent measurements. (I) HT-29 and HCT116 cells were transfected using small interfering RNA against Atg7 for 24 hours, and then treated
with vehicle control or IR-58 for 24 hours, and Atg7 expression was detected using western blot analysis. ( J) Cells were treated as in (I) for 48 hours
and cell viability was measured using CCK-8 kit. β-Actin was taken as the loading control. *p<0.05, **p<0.01, ***p<0.001.
6
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
Figure 5 IR-58 kills tumour cells via inducing excessive autophagy in vivo. Nude mice with human HT-29 subcutaneous tumour xenografts were
established and treated with vehicle control, 24 mg/kg 3-methyladenine (3MA), 3 mg/kg IR-58 with or without 3MA (n=5 per group) every 2 days,
for a total of five treatments. (A) Xenograft size was monitored every day using a sliding calliper, and the volume was estimated using the following
formula: volume=length×width2×0.5. Differences between groups were analysed using a two-way analysis of variance. After excision from the mice,
the xenografts were photographed (B) and weighed (C). (D) The body weight of the mice was measured every day. (E) The xenografts were used for
transmission electron microscopy observations of autophagy and apoptosis. AV, autophagy vacuoles; AP, apoptosis body. (G) Western blot analysis
of autophagy markers (LC3, Atg7 and SQSTM1) of the xenografts. (H) Western blot analysis of apoptosis markers (PARP, c-caspase 3, c-caspase 9)
of the xenografts. (H) H&E staining and immunohistochemical staining of autophagy markers (LC3II, Atg7 and SQSTM1) and apoptosis markers
(c-caspase 3, c-caspase 9) of the xenograft sections. **p<0.01, ***p<0.001. PARP, poly ADP-ribose polymerase; PBS, phosphate buffered saline.
analysis and found that both the activity of Akt and mTOR and
their target genes (p-4EBP1 and p-GSK3β) were inhibited in
IR-58-treated HT-29 and HCT116 cells (figure 6G; p<0.05).
spectrometric analysis using IR-58-treated HT-29 and HCT116
cells. Among the overlapping results of the above analysis,
MTCH2, SLC25A11, TOMM70A, TIMMDC1, TIM44 and
PRDX5 were found to be differentially regulated by IR-58 treat-
ment (figure 7A). Quantitative real-time reverse transcript PCR
(qRT-PCR) was carried out to validate the analysis and demon-
strates the downregulation of TIM44 (figure 7B). TIM44
expression was dose-dependently downregulated by IR-58 at a
transcriptional and translational level (figure 7C, D). TIM44
TIM44 is a novel autophagy-related gene, which is
responsible for IR-58-induced excessive ROS, autophagy and
apoptosis
To demonstrate the upstream components of ROS-mediated
autophagy, we carried out RNA sequencing and mass
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
7

GI cancer
Figure 6 IR-58 induces autophagy through the reactive oxygen species (ROS)-Akt-mammalian target of rapamycin (mTOR) pathway. (A) HT-29 and
HCT116 cells were treated with various doses of IR-58 for 24 hours and ROS production was measured using flow cytometry. (B) A statistical
analysis of the data derived from (A) using a one-way analysis of variance (ANOVA). (C) HT-29 and HCT116 cells were treated with vehicle control,
N-acetylcysteine (NAC), IR-58 with or without NAC for 24 hours and ROS production was measured using flow cytometry. (D) A statistical analysis of
the data derived from (C) using a one-way ANOVA. (E) HT-29 and HCT116 cells were treated as in (C), and cell viability was measured using CCK-8
kit. (F) HT-29 and HCT116 cells were treated as in (C), and expression of the LC3 autophagy marker was measured using western blot analysis. (G)
HT-29 and HCT116 cells were treated as in (A), and phosphorylation of the elements of the Akt-mTOR pathway and their downstream genes
(p-4Ebp 1 and p-GSK3β) was assessed using western blot analysis. β-Actin was taken as the loading control. **p<0.01, ***p<0.001.
expression also decreased in IR-58-treated xenografts in nude
mice (figure 7E). Considering that ROS is mainly generated in
mitochondria and that TIM44 inhibits ROS via inhibition of the
mitochondrial uptake of superoxide dismutase 2 (SOD2, a
major mitochondrial antioxidant enzyme), we hypothesise that
TIM44 inhibition is responsible for IR-58-induced excessive
ROS, autophagy and apoptosis. As shown in figure 7F, mito-
chondrial SOD2 was significantly inhibited by IR-58, which
accounted for the excessive ROS generation. Furthermore,
TIM44 downregulation by siRNA also inhibited mitochondrial
SOD2 expression (figure 7G). TIM44 downregulation also
increased ROS (figure 7H), autophagy (figure 7I) and apoptosis
(figure 7J) in HT-29 and HCT116 cells.
CRC cells compared with corresponding non-tumourigenic
control cells (hepatocyte cell line (L02), non-cancerous lung
bronchial epithelial cell line (HBE) and human non-cancerous
colon epithelium cell line (CCD841CoN)) both at transcrip-
tional (figure 8A) and translational levels (figure 8B). We ana-
lysed human CRC samples and patient histories to investigate
the clinical relevance of TIM44 expression. Western blot ana-
lysis and immunohistochemical staining showed that TIM44
expression levels were higher in CRC tissues compared with
adjacent colon tissues (figure 8C,D). TIM44 expression was
substantially higher in samples from patients with CRC at
the advanced stages (III/IV) in comparison with samples from
those at early stages (I/II; figure 8E, F). Additionally, TIM44
expression was higher in stage M1 compared with stage M0
(figure 8G), suggesting a potential correlation with tumour
metastasis. Although TIM44 expression demonstrated no signifi-
cant change on the basis of gender, age, organ, location, tumour
size and diagnosis, it was much higher in poorly differentiated
CRC with a higher TNM stage (table 1).
TIM44 expression correlates positively with CRC
development and prognosis
As TIM44 was identified as a potential cancer target, we next
investigated the action of TIM44. TIM44 expression was signifi-
cantly higher in lung cancer cells, hepatocellular cancer cells and
8
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
Figure 7 Translocase of inner mitochondrial membrane 44 (TIM44) inhibition is responsible for IR-58-induced excessive reactive oxygen species
(ROS), autophagy and apoptosis. (A) HT-29, HCT116 and A549 cells were treated with vehicle control or 10 μM IR-58 for 24 hours, and then RNA
sequencing and mass spectrometric analysis were performed. The mRNA expression of the overlapped genes is shown. (B) HT-29 cells were treated
as in (A), and MTCH2, SLC25A11, TOM70A, TIMDC1, TIM44 and PRDX5 mRNA expression was assayed using quantitative real-time reverse
transcript PCR (qRT-PCR). (C) HT-29 and HCT116 cells were treated with various doses of IR-58 and TIM44 mRNA expression was measured using
qRT-PCR. (D) HT-29 and HCT116 cells were treated as in (C), and TIM44 protein expression was assayed using western blot analysis. (E) TIM44
expression of xenografts in nude mice treated with vehicle control or IR-58 was stained using immunohistochemistry). (F) HT-29 and HCT116 cells
were treated as in (C), and mitochondrial proteins were extracted for detection of superoxide dismutase 2 (SOD2) using western blot analysis.
Cytochrome c oxidase IV (COX IV) served as the loading control while β-actin served as a negative control. (G) HT-29 and HCT116 cells were
transfected with mock control, small interfering RNA (siRNA) control (siCtrl) or TIM44 siRNA (siTIM44). After 24 hours, mitochondrial proteins were
extracted for measurement of TIM44 and SOD2 expression. (H) HT-29 and HCT116 cells were treated as in (G), and ROS production was measured
using flow cytometry. The results of three independent experiments are shown. (I) HT-29 and HCT116 cells were treated as in (G), and expression of
the LC3 autophagy marker was measured using western blot analysis. ( J) HT-29 and HCT116 cells were treated as in (G), and apoptosis was
measured using flow cytometry. The results of three independent experiments are shown. **p<0.01, ***p<0.001.
Statistical data revealed that patients with CRC with higher
TIM44 expression had poorer overall survival compared with
those with lower TIM44 expression (figure 8H). These findings,
from cell lines to human samples, support the hypothesis that
TIM44 may promote CRC progression.
DISCUSSION
Recent studies have identified that abnormalities in autophagy
might contribute to cancer and some chemotherapeutic agents
show promising anticancer activity through the induction of
autophagy and cell death in various types of cancer, including
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
9

GI cancer
Figure 8 High translocase of inner mitochondrial membrane 44 (TIM44) expression correlates positively with colorectal cancer (CRC) development
and prognosis. TIM44 mRNA (A) and protein (B) expression were measured in lung cancer cells, hepatocellular carcinoma cells and colorectal
carcinoma cells and equivalent non-cancerous cells: lung bronchial epithelial cell line (HBE), hepatocyte cell line (L02) and human colon epithelium
cell line (CCD841CoN). (C) TIM44 expression in human CRC samples and adjacent tissues was measured using western blot analysis. (D)
Immunohistochemical staining of TIM44 in CRC and adjacent tissues. (E) TIM44 expression in non-cancerous tissues, early stage CRC and advanced
stage CRC. (F) Statistical analysis of TIM44 expression levels among CRCs at various stages. (G) Statistical analysis of the correlation between TIM44
expression levels and tumour metastasis. (H) Statistical analysis of the correlation between TIM44 expression levels and overall survival in CRC.
**p<0.01, ***p<0.001.
CRC.3 4 Thus, strategies that augment autophagy are expected to
have potential in treating cancer and the identification of novel
autophagic enhancers is highly anticipated.10 13 Furthermore, the
development of theranostic agents, integrating imaging and
therapeutic modalities, has also received much interest for cancer
treatment.26–28 Therefore, multifunctional agents with simultan-
eous cancer targeting, imaging and killing properties would be of
benefit for personalised oncology. In this study, we have
developed and characterised the NIR small-molecule
fluorophore IR-58, which preferentially accumulates in the
mitochondria of CRC cells, and effectively kills tumour cells
both in vitro and in vivo through the induction of excessive
autophagy. Because NIR fluorescent agents hold great promise
for non-invasive in vivo imaging, as a result of low tissue auto-
fluorescence and deep tissue penetration, the fluorescence prop-
erties of IR-58 could enable real-time tumour visualisation and
may provide a potential agent for tumour-targeted imaging and
therapy in vivo.
ROS is one of the most important regulators of autophagy,
while the mitochondrion is the main source of ROS.29–31 Because
mitochondria are also the key regulators of cell death and mito-
chondrial function is often altered in tumours,32–34 mitochondria-
targeted therapy is a promising anticancer strategy and novel
insights into mitochondrial function are likely to yield new antic-
ancer therapeutic strategies.35 The mitochondrion is composed of
approximately 1500 types of protein, most of which are
synthesised in the cytoplasm and then imported into mitochondria
by protein machinery that is located in the mitochondrial mem-
brane. During this process, TIM44 has been recently identified as
an important player in ROS regulation and the uptake of the
SOD2 nuclear-encoded protein.36 In this study, we demonstrate
for the first time that IR-58 specifically accumulates in mitochon-
dria, and downregulation of TIM44 results in the induction of
autophagy and cell death through the ‘TIM44-SOD2-
ROS-mTOR’ pathway. This work has provided a novel embellish-
ment on the conventional autophagy signalling pathway.
Furthermore, TIM44 is present at much higher levels in cancer
cells compared with non-cancerous cells. Patients with CRC with
lower TIM44 expression live longer than those with higher
TIM44 expression, suggesting that TIM44 may act as a candidate
oncogene and a potential therapeutic target in human CRC.
In summary, we have identified a novel fluorescent therapeutic
autophagy-enhancer, IR-58, which preferentially accumulates in
the mitochondria of tumour cells, inhibits cancer growth and
induces apoptosis via inducing excessive autophagy, which is
mediated through the ROS-Akt-mTOR pathway. Moreover,
TIM44-SOD2-ROS-mTOR signalling inhibition is characterised
for the first time as an autophagy-related pathway, which is
responsible for IR-58-induced excessive ROS and apoptosis (as
illustrated in figure 9). This work may provide a potential strat-
egy to develop small-molecule theranostic agents and holds
promise for cancer targeted imaging and therapy.
10
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
Table 1 Correlations between TIM44 expression and
clinicopathological parameters of CRC
TIM44 expression
Clinicopathological
parameters
Patients
(n)
p
Negative Positive Value
All cases
Gender
151
82
69
0.730
0.131
0.174
0.757
0.743
1.000*
0.000*
Female
63
88
42
61
21
27
Male
Age (years)
<60
84
67
53
50
31
17
≥60
Organ
Colon
52
99
64
39
35
13
Rectal
Location
Left
114
37
77
26
37
11
Right
Tumour size (cm)
<5
69
82
55
48
27
21
≥5
Diagnosis
Adenocarcinoma
Mucinous adenocarcinoma
Histological grade
Well
138
13
94
9
44
4
14
105
32
12
80
11
2
25
21
Moderately
Poorly
Classification
Ulcer
0.019
0.000*
0.000
0.016
0.000
114
37
72
31
42
6
Mass
Figure 9 Proposed model of autophagy-inducing mechanism by
Tumour status
T1+T2
IR-58. IR-58 preferentially accumulates in the mitochondria of tumour
cells, in an organic anion transporter polypeptide (OATP)-dependent
and glycolysis-dependent manner. IR-58 significantly inhibits
24
20
7
4
T3+T4
127
120
Nodal status
N0
translocase of inner mitochondrial membrane 44 (TIM44) expression,
resulting in less uptake of superoxide dismutase 2 (SOD2) into
mitochondria, which leads to excessive reactive oxygen species (ROS)
production and subsequent dephosphorylation of mammalian target of
rapamycin (mTOR), which initiates autophagy. Subsequently,
82
69
67
36
15
33
N1-N2
Metastasis status
M0 (no)
M1 (yes)
TNM stage
I+II
138
13
98
5
40
8
endogenous apoptosis is triggered. PARP, poly ADP-ribose polymerase.
78
73
66
37
12
36
New York, USA) and 1% streptomycin/penicillin (Beyotime,
Shanghai, China). MitoTracker Green was purchased from
Molecular Probes. Hochest 33258 and DAPI were purchased
from Beyotime. 2-DG, oligomycin, chlorpromazine, CCCP, BSP,
3MA and primary antibodies against LC3 (MAP1LC3, western
blot) were purchased from Sigma-Aldrich (St. Louis, Missouri,
USA). Primary antibodies against β-actin, COX IV, LAMP1 and
TIM44 were purchased from Santa Cruz Biotechnology (Santa
Cruz, California, USA). Primary antibodies against LC3II
(immunohistochemistry (IHC)), SQSTM1/P62, Atg7, c-caspase
3, caspase 3, c-caspase 9, p-Akt, t-Akt, p-mTOR, t-mTOR,
p-4EBP1, t-4EBP1 and p-GSK3β were obtained from Cell
Signalling Technology (Boston, Massachusetts, USA). Primary
antibodies against PARP, and SOD2 were obtained from Abcam
III+IV
*Case number <5, corrected with continuity correction of Pearson’s χ2 test.
CRC, colorectal cancer; TIM44, translocase of inner mitochondrial membrane 44;
TNM, tumour, node, metastasis.
MATERIALS AND METHODS
Cell lines and reagents
The human non-cancerous colon epithelial cell line
(CCD841CoN), human colorectal adenocarcinoma cell line
(HT-29), human colorectal carcinoma cell line (HCT116),
human hepatocellular carcinoma cell line (HepG2), human non-
cancerous lung bronchial epithelial cell line (HBE) and human
lung carcinoma cell line (A549) were purchased from the
American Type Culture Collection (Manassas, Virginia, USA).
The human non-cancerous hepatocyte cell line (L02) was pur-
chased from the Chinese Academy of Sciences (Shanghai,
(Cambridge,
Massachusetts,
USA).
Glycine
(98%),
11-bromo-undecane (98%) and 2,3,3-trimethyl-3H indole
(98%) were purchased from ACROS (Beijing, China).
China). Rat transformed dermis-derived mesenchymal cells
37
established and kept in our laboratory.16
All cells were cul-
Human samples
tured in the recommended medium (Hyclone, Logan, Utah,
USA), supplemented with 10% FBS (Gibco, Grand Island,
Human CRC and non-cancerous tissues were obtained from the
Southwest Hospital and Daping Hospital of the Third Military
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
11

GI cancer
Medical University. All tumours were staged by three patholo-
gists, who were blinded to patient information, in the
Department of Pathology of Southwest Hospital, according to
the Union for International Cancer Control classification
system. Informed written consent was provided by all patients.
All human experiments were approved by the Ethics Committee
of the Third Military Medical University.
ROS detection
After IR-58 or siRNA treatment, HT-29 and HCT116 cells
were washed using Dulbecco’s Modified Eagle’s Medium and
incubated with CM-H2DCF-DA for 20 min at 37°C. Cells were
washed and analysed using flow cytometry (BD Biosciences).
siRNA and plasmid transfection
siRNA targeted to human TIM44 and scramble-siRNA were
purchased from Santa Cruz Biotechnology, GFP-LC3 was pur-
chased from Addgene and siRNA targeted to Atg7 was con-
structed by GeneChem (Shanghai, China). HT-29 and HCT116
cells were transfected with siRNA or plasmid using
Lipofectamine 2000 (Invitrogen) in OptiMEM (Hyclone),
according to the manufacturer’s protocol and treated with
IR-58 for another 24 hours.
Synthesis and structural characterisation of IR-58
A
detailed description appears in online supplementary
materials and methods.
Determination of optical properties
The NIR absorption and emission spectra of IR-58 in MeOH
and 10% FBS were determined, and were compared with the
optical properties of ICG under identical conditions.
Ultraviolet-visible-NIR spectroscopy was performed using a
RNA sequencing
Shimadzu
UV-3600
Scanning
Spectrophotometer.
HT-29, HCT116 and A549 cells were treated with vehicle
control or IR-58 for 24 hours and RNA was extracted,
sequenced and analysed by RiboBio (Guangzhou, China).
Photoluminescence was determined using a NIR fluorescence
spectrometer (Lumina Fluorescence Spectrometer, Thermo
Fisher, USA). Emission spectra were obtained using 740 nm
excitation and scanning the wavelength from 750 to 900 nm.
Fluorescence signals were recorded and visually compared using
a NIR imaging system (Kodak In-Vivo FX Professional Imaging
System, New Haven, Connecticut, USA). Based on the fluores-
cence intensity of the emission spectra at the peak wavelength
(807 nm), stability in 10% FBS was also recorded and
compared.
Mass spectrometric analysis
HT-29, HCT116 and A549 cells were treated with vehicle
control or IR-58 for 24 hours and protein was extracted and
analysed by Huada (Guangzhou, China).
Transmission electron microscopy
Cells and mice xenografts were harvested and immediately fixed
in 2.5% glutaraldehyde overnight at 4°C and postfixed with 2%
osmium tetroxide for 1 hour at 37°C. Subsequently, cells were
embedded and stained using uranylacetate/lead citrate. The sec-
tions were imaged using a TEM (JEM-1400PLUS, Japan).
Cell viability assay
Cell proliferation was measured using Cell Counting Kit-8
(Dojindo Laboratories, Kumamoto, Japan) as previously
described.38 HT-29 and HCT116 cells were treated with vehicle
control or various doses of IR-58 for 48 hours, with or without
3MA.
Laser scanning confocal microscopy
HT-29 and HCT116 cells were cultured in a 35 mm petri dish
overnight and incubated with vehicle control or 10 μM IR-58 for
15 min before assessment of the colocalisation of IR-58 and mito-
chondria. Cells were incubated with 150 mM 2-DG, 1 μM oligo-
mycin, 250 μM BSP, 20 μM chlorpromazine, 20 μM CCCP or on
ice for 30 min, and then treated with 10 μM IR-58 for 15 min to
assess IR-58 uptake. Then, cells were washed using phosphate buf-
fered saline (PBS), followed by incubation with MitoTracker
Green (Molecular Probes) for 20 min. Nuclei were stained using
Hoechst. Histological sections from mice and rats treated with
vehicle control or IR-58 were incubated with DAPI for 5 min and
washed with PBS. Cells and tissue sections were imaged using a
confocal fluorescence microscope (Leica, Mannheim, Germany).
Quantitative real-time reverse transcript PCR
Real-time PCR was performed using a SYBR Green qPCR
master mix (Takara) according to the manufacturer’s protocol,
as previously described.38 The primers for the qRT-PCR are
listed in online supplementary table S1.
Western blot analysis
Total proteins were extracted using a cell lysis buffer for western
and IP (Beyotime) and mitochondrial proteins were extracted
using a Cell Mitochondria Isolation Kit (Beyotime). The concen-
trations were determined using a BCA kit (Beyotime). Western
blot analysis was performed as previously described.39
In vivo experiments in nude mice and rats
The in vivo experiments are described in detail in the online
supplementary materials and methods.
Cell cycle assay
A total of 2×106 cells were fixed using 70% ethanol at −20°C,
incubated with propidium iodide (PI) (Invitrogen) for 30 min at
37°C and analysed using fluorescence-activated cell sorting
(FACS) flow cytometry (BD Biosciences, San Jose, California,
USA), as previously described.40
Immunohistochemical staining
Immunohistochemical staining was performed as previously
41
described.38
The detailed information is supplied in the
online supplementary materials and methods.
Apoptosis analysis
The extent of apoptosis was determined by flow cytometry
using an Annexin V/PI staining kit (BD Biosciences), according
to the manufacturer’s instructions. Acidic vesicles were detected
by staining cells with acridine orange at a final concentration of
1.5 mg/mL for 15 min at 37°C. Cells were washed and analysed
using flow cytometry (BD Biosciences).
Statistical analysis
All data are expressed as mean SD, unless otherwise stated.
Comparisons between two groups were performed using a two-
tailed unpaired t-test. Comparisons among three or more
groups were performed using a one-way analysis of variance
(ANOVA). Differences between tumour volumes were
12
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909

GI cancer
16 Tan X, Luo S, Wang D, et al. A NIR heptamethine dye with intrinsic cancer
determined using a two-way ANOVA. Overall survival analysis
was performed using the Kaplan-Meier method. Correlation
between TIM44 expression and clinical parameters was deter-
mined using the Pearson’s χ2 method (with continuity correc-
tion when the number of patients was <5); p<0.05 was
considered to be statistically significant.
targeting, imaging and photosensitizing properties. Biomaterials 2012;33:2230–9.
17 Yue C, Liu P, Zheng M, et al. IR-780 dye loaded tumor targeting theranostic
nanoparticles for NIR imaging and photothermal therapy. Biomaterials
2013;34:6853–61.
18 Zhang E, Luo S, Tan X, et al. Mechanistic study of IR-780 dye as a potential tumor
targeting and drug delivery agent. Biomaterials 2014;35:771–8.
19 Wang Y, Liu T, Zhang E, et al. Preferential accumulation of the near infrared
heptamethine dye IR-780 in the mitochondria of drug-resistant lung cancer cells.
Biomaterials 2014;35:4116–24.
20 Kobayashi H, Longmire MR, Ogawa M, et al. Multiplexed imaging in cancer
diagnosis: applications and future advances. Lancet Oncol 2010;11:589–95.
21 Wang X, Yang L, Chen ZG, et al. Application of nanotechnology in cancer therapy
and imaging. CA Cancer J Clin 2008;58:97–110.
Contributors YH performed the experiments, analysed the data and drafted the
manuscript. JZ and SLL took part in the IHC experiments and analysis of optical
properties. YW, JTH, PL, ZLC, TL and XT participated in the qRT-PCR and western
blot analysis. JZ, JJO, HMM and HJL contributed to collecting human tissues. CMS
designed the study, supervised the experiments and revised the manuscript. All
authors have read and approved the final manuscript.
22 Preston TC, Nuruzzaman M, Jones ND, et al. Role of hydrogen bonding in the
pH-dependent aggregation of colloidal gold particles bearing solution-facing
carboxylic acid groups. J Phys Chem C 2009;113:14236–44.
23 Chong HS, Song HA, Ma X, et al. Bile acid-based polyaminocarboxylate conjugates
as targeted antitumor agents. Chem Commun (Camb) 2009:3011–13.
24 Boni L, David G, Mangano A, et al. Clinical applications of indocyanine
green (ICG) enhanced fluorescence in laparoscopic surgery. Surg Endosc
2015;29:2046–55.
Funding This work was supported by the National Natural Science Foundation of
China (Grant No. 81130026 and 81372727), State Key Basic Research Development
Program (2012CB518103), Program of New Century Excellent Talents in University
(NCET-11-0869) from the Ministry of Education.
Competing interests None.
Patient consent Obtained.
Ethics approval The Ethics Committee of the Third Military Medical University.
Provenance and peer review Not commissioned; externally peer reviewed.
25 Klionsky DJ, Abdelmohsen K, Abe A, et al. Guidelines for the use and interpretation
of assays for monitoring autophagy (3rd edition). Autophagy 2016;12:1–222.
26 Ozdemir V, Williams-Jones B, Glatt SJ, et al. Shifting emphasis from
pharmacogenomics to theragnostics. Nat Biotechnol 2006;24:942–6.
27 Pene F, Courtine E, Cariou A, et al. Toward theragnostics. Crit Care Med 2009;37:
S50–8.
REFERENCES
1
Kim K, Kim JH, Park H, et al. Tumor-homing multifunctional nanoparticles for
cancer theragnosis: simultaneous diagnosis, drug delivery, and therapeutic
monitoring. J Control Release 2010;146:219–27.
28 Chen Z, Ma L, Liu Y, et al. Applications of functionalized fullerenes in tumor
theranostics. Theranostics 2012;2:238–50.
2
3
4
5
Levine B, Kroemer G. Autophagy in the pathogenesis of disease. Cell
2008;132:27–42.
Mizushima N, Levine B, Cuervo AM, et al. Autophagy fights disease through
cellular self-digestion. Nature 2008;451:1069–75.
29 Zhang J, Tripathi DN, Jing J, et al. ATM functions at the peroxisome to induce
pexophagy in response to ROS. Nat Cell Biol 2015;17:1259–69.
30 Chouchani ET, Pell VR, Gaude E, et al. Ischaemic accumulation of succinate controls
reperfusion injury through mitochondrial ROS. Nature 2014;515:431–5.
31 Azad MB, Chen Y, Gibson SB. Regulation of autophagy by reactive oxygen species
(ROS): implications for cancer progression and treatment. Antioxid Redox Signal
2009;11:777–90.
32 Newmeyer DD, Ferguson-Miller S. Mitochondria: releasing power for life and
unleashing the machineries of death. Cell 2003;112:481–90.
33 Gogvadze V, Orrenius S, Zhivotovsky B. Mitochondria in cancer cells: what is so
special about them? Trends Cell Biol 2008;18:165–73.
34 Weinberg SE, Chandel NS. Targeting mitochondria metabolism for cancer therapy.
Nat Chem Biol 2015;11:9–15.
35 Fulda S, Galluzzi L, Kroemer G. Targeting mitochondria for cancer therapy. Nat Rev
Drug Discov 2010;9:447–64.
36 Zhang Y, Wada J, Hashimoto I, et al. Therapeutic approach for diabetic
nephropathy using gene delivery of translocase of inner mitochondrial membrane 44
by reducing mitochondrial superoxide production. J Am Soc Nephrol
2006;17:1090–101.
Sui X, Chen R, Wang Z, et al. Autophagy and chemotherapy resistance: a promising
therapeutic target for cancer treatment. Cell Death Dis 2013;4:e838.
Wang Y, Lin YX, Qiao ZY, et al. Self-assembled autophagy-inducing polymeric
nanoparticles for breast cancer interference in-vivo. Adv Mater Weinheim
2015;27:2627–34.
Law BYK, Chan WK, Xu SW, et al. Natural small-molecule enhancers of autophagy
induce autophagic cell death in apoptosis-defective cells. Sci Rep 2014;4:5510.
Teng ACT, Miyake T, Yokoe S, et al. Metformin increases degradation of
phospholamban via autophagy in cardiomyocytes. Proc Natl Acad Sci USA
2015;112:7165–70.
Tomic T, Botton T, Cerezo M, et al. Metformin inhibits melanoma development
through autophagy and apoptosis mechanisms. Cell Death Dis 2011;2:e199.
Decensi A, Puntoni M, Goodwin P, et al. Metformin and cancer risk in diabetic
patients: a systematic review and meta-analysis. Cancer Prev Res (Phila)
2010;3:1451–61.
6
7
8
9
10 Zhang F, Kumano M, Beraldi E, et al. Clusterin facilitates stress-induced lipidation of
LC3 and autophagosome biogenesis to enhance cancer cell survival. Nat Commun
2014;5:5775.
37 Zhang C, Peng Y, Wang F, et al. A synthetic cantharidin analog for the
enhancement of doxorubicin suppression of stem cell-derived aggressive sarcoma.
Biomaterials 2010;31:9535–43.
11 Rubinsztein DC, Codogno P, Levine B. Autophagy modulation as a potential
therapeutic target for diverse diseases. Nat Rev Drug Discov 2012;11:709–30.
12 Chauhan S, Ahmed Z, Bradfute SB, et al. Pharmaceutical screen identifies novel
target processes for activation of autophagy with a broad translational potential.
Nat Commun 2015;6:8620.
13 Shoji-Kawata S, Sumpter R, Leveno M, et al. Identification of a candidate
therapeutic autophagy-inducing peptide. Nature 2013;494:201–6.
14 Shi C, Zhang C, Su Y, et al. Cyanine dyes in optical imaging of tumours. Lancet
Oncol 2010;11:815–16.
38 Huang Y, Zhou J, He J, et al. SARI, a novel target gene of glucocorticoid receptor,
plays an important role in dexamethasone-mediated killing of B lymphoma cells.
Cancer Lett 2016;373:57–66.
39 Miao H, Ou J, Ma Y, et al. Macrophage CGI-58 deficiency activates
ROS-inflammasome pathway to promote insulin resistance in mice. Cell Rep
2014;7:223–35.
40 Chen HP, Shieh JJ, Chang CC, et al. Metformin decreases hepatocellular carcinoma
risk in a dose-dependent manner: population-based and in vitro studies. Gut
2013;62:606–15.
15 Zhang C, Liu T, Su Y, et al. A near-infrared fluorescent heptamethine indocyanine
dye with preferential tumor accumulation for in vivo imaging. Biomaterials
2010;31:6612–17.
41 Ou J, Miao H, Ma Y, et al. Loss of abhd5 promotes colorectal tumor development
and progression by inducing aerobic glycolysis and epithelial-mesenchymal
transition. Cell Rep 2014;9:1798–811.
Huang Y, et al. Gut 2016;0:1–13. doi:10.1136/gutjnl-2016-311909
13