Induction of mitochondrial apoptosis for cancer therapy: Via dual-targeted cascade-responsive multifunctional micelles

Article abstract of DOI:10.1039/c8tb02159g

Herein, we present a prodrug-loaded multifunctional polymer micelle with hyaluronidase/redox/light multilevel responses and with cell membrane/mitochondrion-dual targeting abilities. This nanocarrier can be internalized by tumor cells via CD44 receptor-mediated targeting. The encapsulated prodrug is released as the carrier is dissociated after the initial degradation of the hyaluronic acid layer by hyaluronidase, followed by the cleavage of the disulfide bonds between hydrophilic and hydrophobic segments in the micelle under the conditions of increased levels of GSH in the cytoplasm. The released prodrug can rapidly target the mitochondria via the TPP function, and convert to the free drug cisplatin through a redox-responsiveness effect. Simultaneously, the membrane permeability of the mitochondria can be improved by the generated reactive oxygen species (ROS) from light irradiation, thus allowing the entry of cisplatin into the mitochondria and causing mitochondrial damage, ultimately leading to mitochondria-mediated apoptosis. Consequently, this nanoformulation shows a highly effective anticancer efficacy in vivo.

Full text of DOI:10.1039/c8tb02159g

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
Induction of mitochondrial apoptosis for cancer
therapy via dual-targeted cascade-responsive
multifunctional micelles†
Cite this: J. Mater. Chem. B, 2018,
6, 8137
Guoqing Wei, Yi Wang, Xuehui Huang, Guang Yang, Jingya Zhao and
Shaobing Zhou
*
Herein, we present
a prodrug-loaded multifunctional polymer micelle with hyaluronidase/redox/light
multilevel responses and with cell membrane/mitochondrion-dual targeting abilities. This nanocarrier can be
internalized by tumor cells via CD44 receptor-mediated targeting. The encapsulated prodrug is released as
the carrier is dissociated after the initial degradation of the hyaluronic acid layer by hyaluronidase, followed
by the cleavage of the disulfide bonds between hydrophilic and hydrophobic segments in the micelle under
the conditions of increased levels of GSH in the cytoplasm. The released prodrug can rapidly target the
mitochondria via the TPP function, and convert to the free drug cisplatin through a redox-responsiveness
effect. Simultaneously, the membrane permeability of the mitochondria can be improved by the generated
reactive oxygen species (ROS) from light irradiation, thus allowing the entry of cisplatin into the
mitochondria and causing mitochondrial damage, ultimately leading to mitochondria-mediated apoptosis.
Consequently, this nanoformulation shows a highly effective anticancer efficacy in vivo.
Received 16th August 2018,
Accepted 9th November 2018
DOI: 10.1039/c8tb02159g
rsc.li/materials-b
of cytochrome C into the cytoplasm, and this would further
activate a cascade of caspase-9 and caspase-3 reactions and
Introduction
Subcellular-targeted therapy has become an important emerging promote tumor cell apoptosis via initiating the mitochondria-
area for cancer treatment with the development of biomedical mediated apoptotic pathways.1,19–21 Therefore, the strategy for
nanotechnology.1–3 As compared to the random interaction of mitochondria-targeted therapy based on nanomedicine is an
drugs in the cell, the therapeutic effect should be enhanced optimal choice for improving antitumor efficiency.
many-fold if the drugs can be specifically directed toward
To achieve precise transport of drugs to mitochondria,
the target organelle (i.e., nucleus,4–6 mitochondria,7,8 or endo- mitochondrial targeting ligand functionalized nanocarriers
plasmic reticulum).9–11 As is commonly known, most of the (or drugs) have received increased attention.22–24 Among these
anticancer drugs, such as camptothecin (CPT), doxorubicin targeting ligands, lipophilic cations—typically triphenylphosphine
(DOX), and cisplatin, have been used in clinical studies for (TPP)—could efficiently attach to mitochondria owing to the
antitumor therapy as DNA-damaging agents.12–15 Unfortunately, increased mitochondrial membrane potential.16,22,25 Nevertheless,
drugs targeting the nuclear genome usually show a low- because the mitochondria have a hydrophobic and dense
therapeutic effect, because the DNA self-repairing pathway is double-membrane system, the latter acts as a formidable barrier
always present in the nucleus.16 In contrast, mitochondrial and prevents many bioactive compounds from gaining entry.
DNA lacks the ability to self-repair, and may be more vulnerable Thus, most mitochondria-targeting prodrugs, or drug-loaded
to damage than nuclear DNA.17 In other words, DNA-damaging mitochondria-targeting nanoparticles, cannot damage the mito-
anticancer agents targeting mitochondrial DNA could present chondria of tumor cells.5,17 Therefore, it is of great importance to
more advantages than drugs targeting the nuclear genome, which provide the nanoparticles or prodrugs with the ability to improve
could enhance the therapeutic effect of anticancer drugs.18 the permeability of the mitochondrial membrane after targeting
Meanwhile, damaging mitochondria could lead to the release the mitochondria. Photosensitizers, for instance Chlorin e6 (Ce6)
(commonly used as a photodynamic therapy (PDT) agent), can
Key Laboratory of Advanced Technologies of Materials, Ministry of Education, and
transfer energy from light irradiation to surrounding dissolved
oxygen (O2) and generate therapeutic reactive oxygen species
(ROS). This results in a significant decrease of mitochondrial
School of Materials Science and Engineering, Southwest Jiaotong University,
Chengdu, Sichuan 610031, P. R. China. E-mail: shaobingzhou@swjtu.edu.cn,
shaobingzhou@hotmail.com
membrane potential (DCm) and a synchronous improvement
of the membrane permeability via oxidizing unsaturated
† Electronic supplementary information (ESI) available: Materials, experimental
details, additional data, Table S1 and Fig. S1–S10. See DOI: 10.1039/c8tb02159g
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8137

View Article Online
Paper
Journal of Materials Chemistry B
Scheme 1 (A) Schematic illustrations of the formation of a PCT-loaded HCCT micelle (PCT@HCCT) via self-assembly and the dissociation of the micelle
through the HAase/GSH sequential response process. (B) Schematic illustration of specifically inducing mitochondrial apoptosis for enhancing cancer
therapy via the micelle: (1) membrane-targeting by receptor-mediated internalization; (2) drug release and mitochondrial targeting: HAase-induced HA
degradation in endosomes and light irradiation-mediated endo/lysosomal escape; mitochondria targeting via TPP function and the improvement of
membrane permeability by generated ROS from light irradiation, resulting in entry of the released cisplatin into mitochondria; (3) cell apoptosis resulting
from the mitochondrial damage due to ROS and cisplatin. (C) The corresponding changes in the chemical structure of polymer HCCT by the sequential
HAase- and redox-responsiveness and prodrug PCT in response to GSH.
phospholipids.8,24 However, the bioactivity of ROS is restricted Moreover, it can be degraded in the presence of hyaluronidase
to an extremely small spherical region centered at the point of its (HAase).28 TPP modified Ce6 is grafted on cystamine-modified
primary generation, owing to their short lifetimes (B40 ns) and HA (HA–CYS) via an amide bond as the hydrophobic block, in
short diffusion range (E20 nm).26,27 To utilize the biological which Ce6 can target mitochondria via the TPP function and
activity of ROS in a more efficient manner and improve antitumor produce ROS in situ under light irradiation (wavelength of
effects, there is an urgent need to develop a new drug delivery 650 nm) to effectively decrease membrane potential (DCm)
system, which can directly carry the therapeutic agent into the and improve the mitochondrial membrane permeability. The
mitochondria to induce mitochondrial apoptosis.
hydrophilic and hydrophobic segments are linked with disulfide
Accordingly, in this study, we developed a newly multi- bonds (SS), which can be cleaved in the presence of high levels of
functional micelle (termed HCCT) with hyaluronidase, redox, glutathione (GSH).29–31 In addition, cisplatin was employed as a
and light multilevel responses, and cell membrane- and model antitumor drug, and it was prepared as prodrug cisplatin
mitochondrion-dual-targeting functions. This micelle can (PCT) that consisted of three distinct components. The first was
encapsulate antineoplastic cisplatin-prodrug (designated as a mitochondria-targeting agent, TPP, which displayed a dramatic
PCT@HCCT) via hydrophobic interactions and p–p stacking enhancement of accumulation of PCT around the mitochondrial
forces during the self-assembly process (Scheme 1A). The matrix membrane. The second is the photosensitizer Ce6 for producing
of the HCCT micelle is the HA–ss–Ce6–TPP polymer, in which ROS under light irradiation. The third is the anticancer drug, which
the hyaluronic acid (HA) that possesses good biocompatibility could induce cellular apoptosis by depolarizing mitochondrion after
as the hydrophilic block can improve micellar internalization its conversion to cisplatin with the endogenous reductants (e.g.,
in tumor cells owing to HA receptor-mediated endocytosis. glutathione, GSH).32,33 Scheme 1B shows that once the PCT-loaded
8138 | J. Mater. Chem. B, 2018, 6, 8137--8147
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry B
Paper
HCCT micelles accumulated in tumor tissue through the EPR effect, between the –NH2 group of the N-Boc-ethylenediamine modi-
they could be internalized by tumor cells via CD44 receptor-mediated fied Pt–COOH (Pt–NH–NH–Boc) and the –COOH group of the
targeting. Subsequently, the HA layer will undergo digestion into TPP modified Ce6 (TPP–Ce6). To improve the biocompatibility
small conjugated pieces by HAase, which has the greatest activity in and enhance the membrane-targeting ability of prodrug PCT,
acid endo/lysosomes.28 Concurrently, light irradiation induces ROS the cystamine modified HA (HA–CYS) provided an –NH2 group
generation, which could disrupt the endo/lysosomal membranes to further conjugate TPP–Ce6 via an acylation reaction to obtain
to release PCT into the cytoplasm,34,35 thus rapidly targeting the a membrane-targeted polymer (HCCT). The latter could be used
mitochondria via the TPP function and allowing conversion to free for loading PCT via hydrophobic interactions and p–p stacking
cisplatin through redox-responsiveness.32,33 At the same time, the forces developed in the process of self-assembly. Furthermore,
membrane permeability of mitochondria would be improved by the the purified intermediates and final products were character-
generated ROS, thus allowing the entry of cisplatin into the mito- ized using 1H nuclear magnetic resonance (1H NMR) or Fourier-
chondria that causes mitochondrial damage, ultimately resulting in transform infrared (FT-IR) spectroscopy (Fig. 1B and C, and
mitochondria-mediated apoptosis. Scheme 1C illustrates the corres- Fig. S1 and S2 in the ESI†). The presence of the peaks for
ponding changes in the chemical structure of the polymer HCCT the characteristic groups of the aromatic ring group of TPP
owing to the sequential HAase and redox-responsive effects and to (7.71–7.80 ppm), the double-bond group of Ce6 (6.16–6.45 ppm),
the chemical structure of the prodrug PCT.
the N-acetyl group of HA (2.0 ppm), and the amidogen group of
Pt–NH3 (6.53 ppm) were confirmed in the 1H NMR spectra
(Fig. 1B and C), thus indicating that HCCT and PCT were
successfully synthesized.
Results and discussion
The blank micelles (HCCT) and PCT-loaded HCCT micelles
The prodrug PCT and the polymer HCCT were synthesized in (PCT@HCCT) were fabricated using a dialysis method. In trans-
accordance with the procedures shown in Fig. 1A. Briefly, the mission electron microscopy (TEM) images (Fig. 2A and B),
prodrug PCT was first synthesized via an acylation reaction both HCCT and PCT@HCCT present a well-dispersed and
Fig. 1 (A) Synthetic route and chemical structure of HCCT conjugate and prodrug PCT. (B) 1H NMR spectrum of HCCT in DMSO-d6: d 7.80–7.71
(m, TPP), 6.18–6.45 (m, –CHQCH2, 3 Â –CHQC2NH–), 2.0 (s, CH3CO–) and (C) PCT in DMSO-d6: d 7.81–7.68 (m, TPP), 6.57–616 (m, 2 Â –NH3,
–CHQCH2, 3 Â –CHQC2NH–), 3.26–3.19 (m, –NH–CH2– and –CH2–NH–), 2.42–2.35 (m, –CO–CH2–CH2–CO– and –CH2–CH2–CO–).
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8139

View Article Online
Paper
Journal of Materials Chemistry B
1
homogeneous core–shell structure. In the H NMR spectra in spectrum, 33.17 wt% of the Ce6 molecules was conjugated
D2O (Fig. 2C), the characteristic peaks of the hydrophobic TPP– to the polymer, which was validated via the grafting rate of
Ce6 block disappeared, however, the peaks of the hydrophilic Ce6 (l = 405 nm) in DMSO (Fig. S4, ESI†). The same results
HA still remained, confirming the core–shell structure of could be calculated using 1H NMR data in Fig. 1B. Interestingly,
PCT@HCCT with TPP–Ce6 in the core and HA within the shell. from the fluorescence spectroscopy results (Fig. 2F), HCCT
After the PCT was loaded, the hydrodynamic size of HCCT and PCT@HCCT showed negligible fluorescence intensity at
slightly increased from 110 to 124 nm, and the zeta potential this excitation wavelength, thus suggesting that the increased
changed from À15 to À8 mV, because HCCT with a negative density of Ce6 in the micelles remained in a self-quenched state
charge was neutralized by the positive charge of PCT (Fig. 2D). from the perspective of fluorescence.38
The size and surface charge of the micelle made it suitable as a
We then investigated the feasibility of HAase to stimulate
nanocarrier for delivering anticancer drugs since it has been fluorescence activation of PCT@HCCT. As displayed in Fig. 2G,
reported that nanoparticles (o200 nm) have a better chance of the fluorescence of PCT@HCCT could be activated by degrad-
accumulating in solid tumors via an enhanced permeability ing the backbone of HA when it was incubated with HAase
and retention (EPR) effect.36,37 The PCT loading content (PCT/ (0.5 mg mLÀ1). Furthermore, the HAase-induced response
HCCT, w/w) improved with the increasing feeding ratio of PCT/ behavior was further investigated based on DLS analysis. As
HCCT (Table S1, ESI†). Based on the optimization, PCT@HCCT shown in Fig. 2H and Fig. S5 (ESI†), when incubated in PBS,
achieved an increased PCT loading content that equalled 15.3 wt% H2O, or RPMI 1640 cell culture medium containing 10% fetal
(PCT/HCCT, w/w), and the feeding PCT : HCCT weight ratio of bovine serum (FBS) without HAase for 7 d, the size of the
0.25 : 1 was used for the following experiments.
PCT@HCCT micelle did not show any change, thus implying
The UV-vis-NIR spectrum revealed the characteristic peaks that the micelle was stable within the blood circulation. How-
of Ce6 (404 and 650 nm) for PCT and HCCT, indicating the ever, when it was incubated with 0.5 mg mLÀ1 HAase for
successful conjugation of Ce6–TPP to HA–CYS. After PCT was 12 h, the micelle size distribution changed from the originally
loaded into HCCT, the characteristic peaks of Ce6 were red- unimodal peak to multimodal peaks (Fig. 2I). The result further
shifted (Fig. 2E). Based on the calculations with the use of the demonstrated that the backbone of HA could have been
Fig. 2 Characterization of the formation and cascade-responsive behaviors of the micelle. (A) TEM image of HCCT micelles and (B) PCT@HCCT
micelles. (C) 1H NMR spectrum of PCT@HCCT in D2O. (D) Particle size and zeta potential of the HCCT and PCT@HCCT micelles. (E) UV-vis spectrum and
fluorescence spectra (F) of free Ce6, PCT, HCCT and PCT@HCCT micelles in DMSO or distilled water. (G) HAase-induced fluorescence emission changes
of PCT@HCCT micelles in phosphate buffer solution (PBS) with HAase at different times. (H) The size distribution of HCCT micelles in PBS at different
times. (I) HAase-induced size changes of HCCT micelles at different times. (J) In vitro PCT release from PCT@HCCT micelles after incubation with
different simulated environments at 37 1C. (K) Relative absorbance profile at 407 nm of DPBF in different samples in PBS solution after irradiation with light
every 10 s. (L) Singlet oxygen quantum yield (FD) of the different samples.
8140 | J. Mater. Chem. B, 2018, 6, 8137--8147
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry B
Paper
degraded by HAase. The final purposes in the development of relative absorbance of DPBF in the PCT@HCCT PBS solution
drug delivery systems in cells include the controlled drug decreased by 18.8% with 650 nm light irradiation within 60 s. The
delivery and its release. Thus, to confirm the enhancement of rapid decrease of absorbance of DPBF suggested that 1O2 was
1
the PCT release from the PCT@HCCT micelle triggered by efficiently produced by PCT@HCCT, and that the O2 quantum
HAase, the released amount of PCT was quantified with induc- yield in the PBS solution was 0.86, which was 1.34-fold higher
tively coupled plasma mass spectrometry (ICP-MS), by detecting than that of free Ce6 (0.64) (Fig. 2L and Fig. S6, ESI†).
the content of Pt using a dialysis bag in a simulated environ-
The intracellular uptake and light-activated ROS generation
ment of the cytoplasm. In Fig. 2J, when HAase was introduced of the micelles were examined in murine B16F10 cells. Since
into the buffer solutions at a pH of 7.4 or 5.0, the PCT release the HA ligand shows an increased binding affinity to the CD44
rate was accelerated compared with the rate at a corresponding receptor, strong interactions between the CD44 receptor on the
pH value without HAase. For instance, the PCT release percen- membranes of B16F10 cells and HA-coordinated ligands are
tages from PCT@HCCT increased from 8% in 1 h to 58% in expected. As shown in Fig. S7B (ESI†), the adsorption behavior
24 h at pH = 7.4 with HAase, and increased from 15% in 1 h to of B16F10 cells to PCT@HCCT was studied using a quartz
72% in 24 h at pH = 5.0 with HAase. The results also indicated that crystal microbalance (QCM) (Q-Sense Explorer, Biolin Scientific,
the outer shell of HA can be degraded into small pieces by HAase, Sweden). Since the mass change on the electrode surface is
which is highly active in acidic tumor microenvironments,28 proportional to the frequency shift of the output oscillation, the
thus leading to a significant reduction in the spatial barrier of frequency shift attributed to interactions between the gold
the micelles.
electrode and cells reached 3.7 Hz after the injection of B16F10
To study the light-triggered singlet oxygen (1O2) generation cells. However, when B16F10 cells were pretreated with molecular
and singlet oxygen quantum yield (FD), 1,3-diphenylisobenzo- HA, the frequency shift increased to 4.7 Hz, thus indicating
furan (DPBF) was selected as an indicator for which absorption a significant enhancement in the cell affinity of PCT@HCCT
was irreversibly quenched by 1O2.39 As shown in Fig. 2K, the by HA coordination. Additionally, B16F10 cell internalization of
Fig. 3 ROS production by PCT@HCCT micelles. (A) CLSM images of PCT@HCCT micelle-induced ROS generation in B16F10 cells under various
conditions. (B) and (C) Flow cytometric examination of intracellular ROS generation of PCT@HCCT induced in B16F10 cells under various conditions.
(D) CLSM images of PCT@HCCT micelle-induced ROS generation in B16F10 cells for different times. (E) and (F) Flow cytometric examination of
intracellular ROS generation of PCT@HCCT induced in B16F10 cells for different times. **p o 0.01 and #p vs. 4 0.05 vs. control (n = 3).
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8141

View Article Online
Paper
Journal of Materials Chemistry B
PCT@HCCT was analyzed via confocal laser scanning microscopy cytotoxicity study, cells exposed to PCT@HCCT showed light-
(CLSM) and flow cytometric examinations. As shown in Fig. S7A, independent apoptotic characteristics. Therefore, it can be
B, and D (ESI†), B16F10 cells incubated with PCT@HCCT with concluded that PCT@HCCT was able to induce more cell
light irradiation displayed remarkably increased fluorescence, apoptosis and death upon light stimulus.
which was approximately 2.2-fold higher than the mean fluores-
To verify whether the prodrug can effectively escape lysosomes,
cence intensity (MFI) in comparison to that without light irradia- lysosomal escape of prodrug in B16F10 cells was visualized
tion. Additionally, when the cells were pretreated with molecular using CLSM. As shown in Fig. S11 (ESI†), after PCT@HCCT
HA, the internalization of PCT@HCCT was decreased. The same is internalized by the cells, hyaluronic acid is degraded by
result can also be found in Fig. S7E (ESI†), which shows the Pt hyaluronidase in lysosomes, which could lead to part of PCT
concentration in cells via ICP-MS detection. These results con- being released from PCT@HCCT and diffusing freely into the
firmed that the internalization of PCT@HCCT can be effectively cytoplasm. In addition, because internalization of PCT@HCCT
improved via CD44 receptor-mediated endocytosis40 and photo- can be effectively improved by photochemical internalization
chemical internalization (PCI).41
(PCI), the red fluorescence of Ce6 is stronger with light
We also investigated the intracellular light-triggered ROS than without light at the same time. Moreover, the reactive
generation of PCT@HCCT using CLSM and flow cytometry. The oxygen, which was produced by Ce6 under light, can destroy the
B16F10 cells were treated with PCT@HCCT with or without lysosome structure, so green fluorescence (lysosomes) was
light irradiation, and the levels of ROS were subsequently weaker with light than without light, thus suggesting that light
measured using ROS detection probes (DCFH-DA). After light irradiation cannot only improve the amount of internalization
irradiation, the bright fluorescence of ROS production (green of PCT@HCCT, but it can also accelerate prodrug escape from
color) in the PCT@HCCT-treated cells was observed by CLSM lysosomes and distribution in the cytoplasm.42 Furthermore,
(Fig. 3A), thus indicating that light induced significant ROS the co-localization analysis of mitochondria was utilized to
generation inside the cells that led to a 5.7-fold increase of MFI observe the targeting of the prodrug to the mitochondria.
compared to that without light irradiation (Fig. 3B and C). Fig. 4A1 and A2 show the CLSM images of the B16F10 cells
However, ROS generation was suppressed when the cells were treated with PCT@HCCT with light irradiation (10 min) after
pre-incubated with ROS scavenger vitamin C (Vc). Additionally, incubation for 1 and 6 h, respectively. The corresponding fluores-
the fluorescence intensity increased in a time-dependent manner cence intensity profiles across the white line (marked in the CLSM
(Fig. 3D), which is approximately equivalent to a 2.5-fold increase images) are shown in Fig. 4B1 and B2. Quantitative analysis of the
of MFI after the B16F10 cells were treated with PCT@HCCT for fluorescence intensity and subcellular co-localization analysis
6 h compared to 1 h treatment (Fig. 3E and F). These results were derived from these fluorescence images. Pearson’s correla-
demonstrated that the ROS can be generated only under light tion coefficient (Rr) was utilized to quantitatively evaluate the
irradiation.
correlation between the two types of fluorescence co-localization
We then examined the cytotoxicity of PCT@HCCT on B16F10 analysis.43 According to this, we can quantitatively determine
cells using live/dead staining with the Alamar blue assay. PBS with whether the prodrug is able to target the mitochondria effectively
light irradiation was used as the control. Fig. S9 (ESI†) shows that after the cells were incubated with PCT@HCCT for 1 or 6 h. As
the cell viability was 490% for both HUVEC cells (normal cells) and shown in Fig. 4C, the mean Rr values of the red (Ce6) and the
B16F10 (tumor cells) after incubation with the nanocarriers (HCCT) green fluorescence (MitoTracker green) at 1 and 6 h were 0.87 and
with concentrations in the range 10–250 mg mLÀ1 for 24 h, thus 0.92, respectively, thus indicating that there were no obvious
indicating that the blank micelles had excellent cytocompatibility. differences of co-localization after incubation for 1 and 6 h. The
Fig. S10A (ESI†) shows that PCT@HCCT with light irradiation was results also illustrated that the prodrug that escaped from lyso-
more toxic (IC50 of PCT is 0.421 mM) compared to that without light somes could rapidly accumulate around the mitochondria.
irradiation (IC50 of PCT is 1.63 mM), but the toxicity effect was
The TEM images in Fig. 4D1–D3 and D5–D7 show the intra-
weakened when the cells were pre-incubated with the ROS scavenger cellular transport process of cisplatin in PCT@HCCT in B16F10 cells
Vc. This difference is attributed to the ROS generation by Ce6 in at 6 h. As shown in the magnified images, cisplatin could enter the
PCT@HCCT under light irradiation. Live/dead staining was mitochondria in the B16F10 cells after incubation with PCT@HCCT
performed to further evaluate the light-dependent cytotoxicity, and following light irradiation. However, no cisplatin could be
and the result was consistent with the Alamar blue assay since found‘ in the mitochondria of the B16F10 cells after incubation
more cells died when they were treated with PCT@HCCT under with PCT@HCCT but without the use of light irradiation. This
light irradiation (Fig. S10B, ESI†).
result further confirms that chemotherapeutic drugs struggle to
An annexin V-FITC/propidium iodide (PI) double-staining pass through the complete mitochondrial double membrane
assay was used to evaluate the apoptosis in B16F10 cells structure.17 In addition, the presence of platinum was confirmed
induced by PCT@HCCT under different processing conditions. based on energy dispersive spectrometry (EDS) analysis of the two
Compared to the treatment without light irradiation (a total groups (Fig. 4D4 and D8). Additionally, only platinum was detected
apoptotic ratio of 53.03%), the percentage of B16F10 cells in the mitochondria of B16F10 cells in the case of PCT@HCCT in
undergoing apoptosis after incubation with PCT@HCCT the light irradiation group, thus illustrating that cisplatin entered
with light irradiation (a total apoptotic ratio of 83.24%) was mitochondria after the membrane permeability was improved
significantly increased (Fig. S10C and D, ESI†). Similar to the using in situ ROS generation following light irradiation.
8142 | J. Mater. Chem. B, 2018, 6, 8137--8147
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 4 Mitochondrial co-localization analysis. (A1) Confocal images of B16F10 cells after incubation with PCT@HCCT micelles with irradiation (10 min)
for (A1 or A2) 6 h. (B1 and B2) Profiles of the distribution of the fluorescence signals of Ce6 (Red) and MitoTracker Green along the signified white line in
the confocal microscopic images. (C) Co-localization ratio between Ce6 and MitoTracker Green signified by the Rr value after the B16F10 cells were
incubated with PCT@HCCT micelles with irradiation (10 min) for 1 and 6 h. (D1–D3) and (D5–D7) Mitochondria ultrastructure observed by using TEM.
(D1–D3) B16F10 cells were incubated with PCT@HCCT micelles without irradiation and (D5–D7) with irradiation (10 min). (D4 and D8) Elemental analysis
of mitochondria. N: nucleus; M: mitochondria. In (A1 and A2), 1 and 5: red fluorescence of Ce6, 2 and 6: green fluorescence of #p 4 0.05 (n = 5).
To further investigate the mechanism of how PCT@HCCT in a noticeable increase of the green fluorescence intensity due
the presence of light irradiation induced cell apoptosis, we to the collapse of the mitochondrial DCm. The intensity of
systematically studied the changes in mitochondrial membrane the green fluorescence further decreased when the cells were
potential and the expression of related proteins in B16F10 pre-incubated with a ROS scavenger Vc. Quantitative analysis of
following light irradiation. Fluorescent probe JC-1 staining was the ratio between the green-to-red fluorescence using flow cyto-
used to monitor the change in mitochondrial membrane metric measurements confirmed that PCT@HCCT significantly
potential (DCm).44 When the permeability of the mitochondrial caused mitochondrial depolarization (Fig. 5B and C). This result
membrane increases, DCm will decrease.45 When the mitochon- indicates that the prodrug can localize in the mitochondria, and
dria are normal with an intact mitochondrial membrane system, the local ROS of the organelles subsequently increase rapidly
JC-1 will display potential-dependent accumulation in the mito- with light irradiation, further leading to a rapid decrease in DCm
chondrial matrix, which is accompanied by red fluorescence owing to the fact that the mitochondria are sensitive to ROS.47
emission owing to the formation of J-aggregates, thus indicating
In comparison, the expression level of proteins was further
an increased DCm. In contrast, depolarized mitochondria dis- investigated by Western blot assays (Fig. 5D) in order to
play green fluorescence owing to the formation of J-monomers in evaluate whether the mitochondrial damage and the activation
the cytoplasm, thereby indicating a lower DCm.46 Therefore, the of the mitochondrial apoptotic pathway can be caused by the
change in membrane potential can be evaluated by measuring entry of cisplatin in the mitochondria and its subsequent
the green/red fluorescence intensity ratio. Fig. 5A demonstrates action on mitochondrial DNA. It is known that the damage of
that PCT@HCCT without light irradiation has a negligible mitochondria could cause a remarkable release of cytochrome C
influence on the mitochondrial membrane potential. The mito- into the cytosol that would prompt cell apoptosis.48,49 Additionally,
chondria were depolarized with light irradiation as indicated by the improvement of the membrane permeability of mitochondria
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8143

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 5 Mitochondrial membrane potential (DCm) analysis. (A) CLSM images of DCm shift as determined by JC-1 assay of B16F10 cells. (B) Flow cytometric
curves of JC-1 distribution in B16F10 cells. (C) The JC-1 monomer to aggregate (green/red) ratio in the cells as determined by flow cytometry. **p o 0.01
and #p vs. 40.05 vs. control (n = 3). (D) Representative cytochrome C and caspase-3 protein were analyzed by western blot. Glyceraldehyde
3-phosphate dehydrogenase (GAPDH) was used as a loading control. (E and F) Cytochrome C and caspase-3 protein relative to GAPDH were determined
by Image J, respectively. **p o 0.01 denote statistically significant differences.
further promoted the release of cytochrome C into the cytoplasm.
To further assess the in vivo anticancer efficacy of
As shown in Fig. 5E, the cytochrome C level of PCT@HCCT was PCT@HCCT, various formulations were intravenously injected
increased with light irradiation, but it decreased when the cells into B16F10 tumor-bearing mice every 2 days three times,
were pre-incubated with the ROS scavenger Vc. Meanwhile, the including days 0, 3, and 6. The light source for therapy was a
expression of caspase-3 and cleaved caspase-3, which is a major 650 nm laser producing a photodensity of 250 mW cmÀ2. As
cell death effector protease, was also determined. The caspase-3 shown in Fig. 6B and C, there is an effective inhibition of tumor
expression and cleaved caspase-3 level of PCT@HCCT with light growth in mice treated with PCT@HCCT using light irradia-
irradiation were significantly downregulated and upregulated in tion. At day 15, PCT@HCCT without light irradiation and
contrast to that without light irradiation, respectively (Fig. 5F and PCT@mPEG–PCL with light irradiation displayed moderate
Fig. S14B, ESI†), thus suggesting the occurrence of mitochondrial anticancer capability with the tumor volumes of 1.68 and
damage and the initiation of the mitochondria-mediated 1.24 cm3, respectively. However, PCT@HCCT with light irradiation
apoptotic pathway.
displayed significantly higher inhibitory effects on tumor growth,
To explore the in vivo tumor-targeted capability of PCT@HCCT, and the tumor volume was only 0.51 cm3. The inhibition rate of the
B16F10 tumor bearing mice were used as a tumor xenograft tumor growth in the PCT@HCCT and light irradiation group is
model. Compared to the noncellular membrane-targeted group 84.4%, which is 1.7-fold higher than that in the PCT@HCCT
(PCT@mPEG–PCL), a remarkably increased fluorescence Ce6 group without the use of light irradiation (49.6%), and 1.3-fold
signal was observed in the tumor region at 1, 6, and 12 h post- higher than that in the PCT@mPEG–PCL and light irradiation
injection of PCT@HCCT (Fig. 6A). The tumor and normal group (62.8%). The therapeutic efficacy of PCT@HCCT was
tissues were collected for ex vivo imaging at 12 h, and the further evaluated by histological examination. The images of
relative fluorescence signal of Ce6 was mainly distributed in the hematoxylin and eosin (HE) stained tumor tissue treated by
tumor and liver tissues (Fig. 6A). These results indicated that PCT@HCCT with light irradiation indicated the presence of a
PCT@HCCT could accumulate at the tumor site via both the large range of dead cells without nuclei (Fig. 6F). Additionally,
EPR effect and HA receptor-mediated endocytosis.
deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL)
8144 | J. Mater. Chem. B, 2018, 6, 8137--8147
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 6 In vivo antitumor therapy. (A) In vivo fluorescence images of B16F10 tumor bearing nude mice after intravenous injection of PCT@HCCT and
PCT@mPEG–PCL, and ex vivo fluorescence images of isolated tissues at 12 h post-injection. (B) Excised B16F10 tumors (15th d after the first treatment)
from tumor-bearing C57BL/6 mice subjected to different treatments. (C) Tumor size, (D) body weight changes, and (E) survival curves. (F) Ex vivo
histological analyses of tumor sections. In hematoxylin and eosin-stained sections, nuclei are stained blue and the extracellular matrix and cytoplasm are
stained red. In terminal deoxynucleotidyl transferase dUTP nick end labeling analysis, blue and brown stains indicate normal and apoptotic
cells, respectively. In (B–F), (1): PBS + L; (2): HCCT; (3): PCT@HCCT; (4): PCT@mPEG–PCL + L; (5): PCT@HCCT + L. **p o 0.01 denote statistically
significant differences.
staining (Fig. 6F) was also employed to evaluate the tumor cell cancer therapy because it possesses both high antitumor effi-
apoptosis caused by various formulations, where the group cacy and low-systemic toxicity.
treated with PCT@HCCT and light appeared to have the
highest level of cell apoptosis in tumor tissues. Collectively,
these results verified that the effect of antitumor therapy of
PCT@HCCT can be significantly improved under light irradia-
Conclusions
tion, owing to the capabilities of targeted mitochondria therapy In summary, we successfully developed a new, dual-targeted,
with nanocarriers.
cascade-responsive multifunctional micelle to specifically
To evaluate the in vivo safety of the nanocarrier, the mice induce mitochondrial apoptosis in cancer therapy. The drug
body weight, survival rate, and histological analyses of organs loading content of this micelle was significantly enhanced via
were assessed. All groups showed no significant change in mice hydrophobic interactions and p–p stacking forces developed
body weight during the treatment (Fig. 6D). Furthermore, the between the micellar matrix and the drug molecules. The
survival rate of mice in the PCT@HCCT and light group was the micelle could be selectively internalized by tumor cells through
highest (85.7%) among all groups during a period of 45 d membrane-targeting, and the HA layer could be subsequently
(Fig. 6E), thus implying that this was the safest treatment. degraded into small conjugated pieces by hyaluronidase. Con-
General body toxicity was further evaluated by HE staining of currently, the ROS generated from light irradiation disrupted
the major organs (e.g., heart, liver, spleen, lung, and kidney). the endo/lysosomal membranes to prompt the release of PCT,
The HE-stained images in Fig. S15 (ESI†) show that there were which possessed a considerable capacity in targeting the mito-
no appreciable histological changes and undetectable adverse chondria owing to the TPP function and conversion into free
effects (e.g., inflammatory response or necrosis) in the main cisplatin through the redox-responsive effect in the cytoplasm.
organs, thus indicating excellent biocompatibility in vivo. Simultaneously, the membrane permeability of the mitochon-
Therefore, the micelle is an excellent nanoformulation for dria could be effectively improved by the generated ROS, thus
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8145

View Article Online
Paper
Journal of Materials Chemistry B
15 Z. T. Cao, Z. Y. Chen, C. Y. Sun, H. J. Li, H. X. Wang,
Q. Q. Cheng, Z. Q. Zuo, J. L. Wang, Y. Z. Liu, Y. C. Wang and
J. Wang, Biomaterials, 2016, 94, 9.
16 S. Marrachea, R. K. Pathaka and S. Dhar, Proc. Natl. Acad.
Sci. U. S. A., 2014, 111, 10444.
17 S. R. Jean, M. Ahmed, E. K. Lei, S. P. Wisnovsky and
S. O. Kelley, Acc. Chem. Res., 2016, 49, 1893.
18 Y. Wang, G. Wei, X. Zhang, X. Huang, J. Zhao, X. Guo and
S. Zhou, Small, 2018, 14, 1702994.
allowing cisplatin to enter the mitochondria and cause
mitochondrial damage, ultimately resulting in mitochondrial
apoptosis. As a result, the micelle represented a highly effective
anticancer efficacy with a tumor growth inhibition rate of
84.4% and a survival rate of 85.7% in mice. Therefore, the
dual-targeted and cascade-responsive multifunctional micelles
are envisaged to be a promising nanocarrier for mitochondrial
therapy of malignant tumors.
19 J. Xu, F. Zeng, H. Wu, C. Hu and S. Wu, Biomacromolecules,
2014, 15, 4249.
Conflicts of interest
20 S. Biswas and V. P. Torchilin, Adv. Drug Delivery Rev., 2014,
66, 26.
The authors declare no competing financial interest.
21 B. F. Zhang, L. Xing, P. F. Cui, F. Z. Wang, R. L. Xie,
J. L. Zhang, M. Zhang, Y. J. He, J. Y. Lyu, J. B. Qiao,
B. A. Chen and H. L. Jiang, Biomaterials, 2015, 61, 178.
22 Y. Tan, Y. Zhu, Y. Zhao, L. Wen, T. Meng, X. Liu,
X. Yang, S. Dai, H. Yuan and F. Hu, Biomaterials, 2018,
154, 169.
23 G. Yu, D. Wu, Y. Li, Z. Zhang, L. Shao, J. Zhou, Q. Hu,
G. Tang and F. Huang, Chem. Sci., 2016, 7, 3017.
24 B. Feng, F. Zhou, Z. Xu, T. Wang, D. Wang, J. Liu, Y. Fu,
Q. Yin, Z. Zhang, H. Yu and Y. Li, Adv. Funct. Mater., 2016,
26, 7431.
25 R. A. J. Smith, C. M. Porteous, A. M. Gane and M. P. Murphy,
Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 5407.
26 M. Ethirajan, Y. Chen, P. Joshi and R. K. Pandey, Chem. Soc.
Rev., 2011, 40, 340.
27 Z. Yu, Q. Sun, W. Pan and Na. Li and B. Tang, ACS Nano,
2015, 9, 11064.
Acknowledgements
This work was partially supported by the China National Funds
for Distinguished Young Scientists (51725303), the National
Natural Science Foundation of China (21574105, 81701831, and
51703189), the Sichuan Province Youth Science and Technology
Innovation Team (2016TD0026) and Doctoral Innovation Fund
Program of Southwest Jiaotong University (D-CX201730).
References
1 S. Fulda, L. Galluzzi and G. Kroeme, Nat. Rev. Drug Discovery,
2010, 9, 447.
2 B. P. Shah, N. Pasquale, G. De, T. Tan, J. Ma and K.-B. Lee,
ACS Nano, 2014, 8, 9379.
3 Y. Zhang, Y. Shen, X. Teng, M. Yan, H. Bi and P. C. Morais,
ACS Appl. Mater. Interfaces, 2015, 7, 10201.
4 X. Guo, X. Wei, Y. Jing and S. Zhou, Adv. Mater., 2015,
27, 6450.
28 R. Stern and M. J. Jedrzejas, Chem. Rev., 2006, 106, 818.
29 X. Guo, C. Shi, G. Yang, J. Wang, Z. Cai and S. Zhou, Chem.
Mater., 2014, 26, 4405.
30 C. Shi, X. Guo, Q. Qu, Z. Tang, Y. Wang and S. Zhou,
Biomaterials, 2014, 35, 8711.
31 J. Wang, G. Yang, X. Guo, Z. Tang, Z. Zhong and S. Zhou,
Biomaterials, 2014, 35, 3080.
32 Y. R. Zheng, K. Suntharalingam, T. C. Johnstone, H. Yoo,
W. Lin, J. G. Brooks and S. J. Lippard, J. Am. Chem. Soc.,
2014, 136, 8790.
33 P. Ma, H. Xiao, C. Yu, J. Liu, Z. Cheng, H. Song,
X. Zhang, C. Li, J. Wang, Z. Gu and J. Lin, Nano Lett.,
2017, 17, 928.
5 K. Han, W. Y. Zhang, J. Zhang, Q. Lei, S. B. Wang, J. W. Liu,
X. Z. Zhang and H. Y. Han, Adv. Funct. Mater., 2016,
26, 4351.
6 J. Wang, S. Shen, D. Li, C. Zhan, Y. Yuan and X. Yang, Adv.
Funct. Mater., 2018, 28, 1704806.
7 K. Han, J. Y. Zhu, H. Z. Jia, S. B. Wang, S. Y. Li, X. Z. Zhang
and H. Y. Han, ACS Appl. Mater. Interfaces, 2016, 8, 25060.
8 K. Han, Q. Lei, S. B. Wang, J. J. Hu, W. X. Qiu, J. Y. Zhu,
W. N. Yin, X. Luo and X. Z. Zhang, Adv. Funct. Mater., 2015,
25, 2961.
34 C. Qian, P. Feng, J. Yu, Y. Chen, Q. Hu, W. Sun, X. Xiao,
X. Hu, A. Bellotti, Q. D. Shen and Z. Gu, Angew. Chem.,
Int. Ed., 2017, 56, 2588.
35 H. He, S. Ji, Y. He, A. Zhu, Y. Zou, Y. Deng, H. Ke, H. Yang,
Y. Zhao, Z. Guo and H. Chen, Adv. Mater., 2017, 29, 1606690.
36 H. Maeda, J. Wu, T. Sawa, Y. Matsumura and K. Hori,
J. Controlled Release, 2000, 65, 271.
9 H. C. Wang, S. C. Hsieh, J. H. Yang, S. Y. Lin and L. Y. Sheen,
Nutr. Cancer, 2012, 64, 770.
10 K. S. Eom, H. J. Kim, H. S. So, R. Park and T. Y. Kim, Biol.
Pharm. Bull., 2010, 33, 1644.
11 S. Zhu, J. Jin, Y. Wang, Z. Ouyang, C. Xi, J. Li, Y. Qiu, J. Wan,
M. Huang and Z. Huang, Food Chem. Toxicol., 2012,
50, 1149.
37 G. Wei, Y. Wang, X. Huang, H. Hou and S. Zhou, Small
Methods, 2018, 1700358.
12 X. Hu, S. Zhai, G. Liu, D. Xing, H. Liang and S. Liu, Adv.
Mater., 2018, 30, 1706307.
38 L. Feng, D. Tao, Z. Dong, Q. Chen, Y. Chao, Z. Liu and
M. Chen, Biomaterials, 2017, 127, 13.
13 X. Guo, L. Wang, K. Duval, J. Fan, S. Zhou and Z. Chen, Adv.
Mater., 2018, 30, 1705436.
39 X. Zhen, C. Zhang, C. Xie, Q. Miao, K. L. Lim and K. Pu, ACS
Nano, 2016, 10, 6400.
14 Y. Wang, L. Zhang, X. Zhang, X. Wei, Z. Tang and S. Zhou,
ACS Appl. Mater. Interfaces, 2016, 8, 5833.
8146 | J. Mater. Chem. B, 2018, 6, 8137--8147
This journal is ©The Royal Society of Chemistry 2018

View Article Online
Journal of Materials Chemistry B
Paper
40 Y. Zhong, J. Zhang, R. Cheng, C. Deng, F. Meng, F. Xie and 46 M. Okadaa, N. I. Smithb, A. F. Palonponc, H. Endoa,
Z. Zhong, J. Controlled Release, 2015, 205, 144.
41 B. S. J. M. I. Touhid, J. Cancer Res. Exp. Oncol., 2011, 3, 26.
S. Kawataa, M. Sodeokac and K. Fujita, Proc. Natl. Acad.
Sci. U. S. A., 2012, 109, 28.
42 Y. Wang, T. Yang, H. Ke, A. Zhu, Y. Wang, J. Wang, J. Shen, 47 H. S. Jung, J. Han, J. H. Lee, J. H. Lee, J. M. Choi,
G. Liu, C. Chen, Y. Zhao and H. Chen, Adv. Mater., 2015,
27, 3874.
H. S. Kweon, J. H. Han, J. H. Kim, K. M. Byun, J. H. Jung,
C. Kang and J. S. Kim, J. Am. Chem. Soc., 2015, 137, 3017.
43 Y. Wang, G. Wei, X. Zhang, F. Xu, X. Xiong and S. Zhou, Adv. 48 M. E. Guicciardi, J. Deussing, H. Miyoshi, S. F. Bronk,
Mater., 2017, 29, 1605357.
44 B. N. Desai, B. R. Myers and S. L. Schreiber, Proc. Natl. Acad.
Sci. U. S. A., 2002, 99, 4319.
P. A. Svingen, C. Peters, S. H. Kaufmann and G. J. Gores,
J. Clin. Invest., 2000, 106, 1127.
49 J. R. J. Caruso, Jr, P. Mathieu, B. Chelladurai, X. M. Yin and
D. Kessel, Cell Death Differ., 2002, 9, 934.
45 G. Kroemer and J. C. Reed, Nat. Med., 2000, 6, 513.
This journal is ©The Royal Society of Chemistry 2018
J. Mater. Chem. B, 2018, 6, 8137--8147 | 8147