Biomaterials 277 (2021) 121061

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Enhanced chemo-photodynamic therapy of an enzyme-responsive prodrug

in bladder cancer patient-derived xenograft models
Ping Tan a, 1, Hao Cai a, 1, Qiang Wei a, Xiaodi Tang c, Qianfeng Zhang a, Michal Kopytynski f,
Junxiao Yang d, Yong Yi c, Hu Zhang e, Qiyong Gong a, b, Zhongwei Gu a, Rongjun Chen f,
Kui Luo a, *
a
Department of Urology, Institute of Urology, Huaxi MR Research Center (HMRRC), National Clinical Research Center for Geriatrics, Department of Radiology,
Functional and Molecular Imaging Key Laboratory of Sichuan Province, West China Hospital, Sichuan University, Chengdu, 610041, China
b
Research Unit of Psychoradiology, Chinese Academy of Medical Sciences, Chengdu, 610041, China
c
State Key Laboratory for Environment-friendly Energy Materials, Southwest University of Science and Technology, Mianyang, 621010, China
d
State Key Laboratory Cultivation Base for Nonmetal Composite and Functional Materials, Southwest University of Science and Technology, Mianyang, 621010, China
e
Amgen Bioprocessing Centre, Keck Graduate Institute, CA, 91711, USA
f
Department of Chemical Engineering, Imperial College London, South Kensington Campus, London, SW7 2AZ, United Kingdom

A R T I C L E I N F O A B S T R A C T

Keywords: Patient-derived xenograft (PDX) models are powerful tools for understanding cancer biology and drug discovery.
Polymer-paclitaxel prodrug In this study, a polymeric nano-sized drug delivery system poly (OEGMA)-PTX@Ce6 (NPs@Ce6) composed of a
Chemo-photodynamic therapy photosensitizer chlorin e6 (Ce6) and a cathepsin B-sensitive polymer-paclitaxel (PTX) prodrug was constructed.
Photo-chemical internalization
The photochemical internalization (PCI) effect and enhanced chemo-photodynamic therapy (PDT) were achieved
PDX models
via a two-stage light irradiation strategy. The results showed that the NPs@Ce6 had great tumor targeting and
Bladder cancer
Bioinformatics analysis rapid cellular uptake induced by PCI, thereby producing excellent anti-tumor effects on human bladder cancer
PDX models with tumor growth inhibition greater than 98%. Bioinformatics analysis revealed that the combi­
nation of PTX chemotherapy and PDT up-regulated oxidative phosphorylation and reactive oxygen species (ROS)
generation, blocked cell cycle and proliferation, and down-regulated the pathways related to tumor progression,
invasion and metastasis, including hypoxia, TGF-β signaling and TNF-α signaling pathways. Western blots
analysis confirmed that proteins promoting apoptosis (Bax, Cleaved caspase-3, Cleaved PARP) and DNA damage
(γH2A.X) were up-regulated, while those inhibiting apoptosis (Bcl-2) and mitosis (pan-actin and α/β-tubulin)
were down-regulated after chemo-PDT treatment. Therefore, this stimuli-responsive polymer-PTX prodrug-based
nanomedicine with combinational chemotherapy and PDT evaluated in the PDX models could be a potential
candidate for bladder cancer therapy.

1. Introduction premature release of therapeutic agents encapsulated in the nanocarrier

into the blood system, and achieving deep penetration into the tumor. In
Nanomedicines produced by the incorporation of therapeutic agents addition, after overcoming these barriers, the nanomedicine must be
in a nanoscale polymer drug delivery system to treat malignant tumors able to be effectively engulfed by tumor cells, and the encapsulated
have made considerable progress, however, their therapeutic effects are drugs must be released specifically and rapidly inside tumor cells [3–6].
still unsatisfactory due to the complexity of the macroenvironment of All these steps play a critical role in maximizing therapeutic effects of
the human body and the tumor microenvironment [1,2]. The nano­ nanomedicines. Therefore, tumor microenvironment-responsive poly­
medicine needs to overcome multiple physiological barriers before mer-based nanomedicines have been designed to specifically overcome
arriving at the tumor site after intravenous administration, including these biological barriers with improved delivery efficiencies to tumor
evading elimination by the mononuclear phagocyte system, avoiding tissues, resulting in enhanced therapeutic indexes as well as reduced

* Corresponding author.
E-mail address: [email protected] (K. Luo).
1
Dr. Ping Tan and Dr. Hao Cai contributed equally to this work.

https://doi.org/10.1016/j.biomaterials.2021.121061
Received 24 June 2021; Received in revised form 3 August 2021; Accepted 4 August 2021
Available online 27 August 2021
0142-9612/© 2021 Elsevier Ltd. All rights reserved.
P. Tan et al. Biomaterials 277 (2021) 121061

adverse effects [7,8]. maintaining the function of PEGylation and enhancing cellular uptake of
To prepare such polymeric prodrug-based nanomedicines with high nanoparticles should be carefully considered for poly (OEGMA)-based or
stability during circulation and rapid drug release at the tumor site, the PEGylated nanomedicines.
properties of polymeric carriers need to be finely tuned, including their Various methods have been used to enhance the cellular uptake of
chemical composition, chemical structure and molecular weight [9]. An nanoparticles including grafting of targeting moieties on nanoparticles
outer layer of polyethylene glycol (PEG) on nanoparticles can signifi­ and tuning the morphology and/or size of nanoparticles [18–20]. A new
cantly reduce non-specific interactions between nanoparticles and cells strategy of photochemical internalization (PCI) has been developed by
and decrease adsorption of serum proteins during the circulation, encapsulating a photosensitizer (PS) in nanoparticles. In such systems,
therefore, PEGylation of nanoparticles has become a commonly used the permeability of cell membranes is increased to facilitate the cellular
method to overcome the elimination of nanoparticles by the mono­ uptake of nanoparticles due to the effect of reactive oxygen species
nuclear phagocyte system [10,11]. Recently, poly (oligo (ethylene gly­ (ROS) generated by PS under transitory low-energy-density irradiation
col) methacrylate) (poly (OEGMA)) has been reported as an efficient [21–23].
drug delivery carrier due to the synergistic effect of PEGylation, its high In addition, the nanomedicine incorporating a PS could be used for a
molecular weight and flexibility for chemical modification [12–15]. The combinational therapy since the PS is also an essential component for
density of PEG or oligo (ethylene glycol) (OEG) has a great impact on the photodynamic therapy (PDT) [24,25]. It has been demonstrated that a
uptake of nanoparticles by cells [16,17]. Therefore, a balance of single chemotherapeutic drug in a nanomedicine can be ineffective in

Scheme 1. Schematic illustration of (a) con­

structing a bladder cancer PDX model; (b)
degradation of the poly (OEGMA)-based polymer
prodrug upon treatment with cathepsin B to
release PTX; and (c) self assembly of the amphi­
philic polymer poly (OEGMA)-PTX into nano­
particles NPs@Ce6 encapsulated with the
hydrophobic Ce6. (d) NPs@Ce6 can produce ROS
and exert the PCI effect under light irradiation
for a short time and enhance the cellular uptake
of NPs@Ce6. The PTX and Ce6 are released from
NPs@Ce6 via Cathepsin B-responsive degrada­
tion in a tumor cell. Under long-time light irra­
diation, Ce6 can produce a large amount of ROS,
which has a synergistic anti-tumor effect with
PTX-mediated chemotherapy, and induce cell
death by blocking cell mitosis, promoting cell
apoptosis and damaging DNA.

2
P. Tan et al. Biomaterials 277 (2021) 121061

complete elimination of tumors and it often induces chemo-resistance 2.3. Human tumor samples and PDX mice model
from tumors during the treatment [26]. Combining chemotherapy and
PDT has become a viable option to improve the effectiveness of cancer The study protocol and subsequent amendments were approved by
treatment [27]. PDT relies on cytotoxic ROS, especially singlet oxygen, the Institutional Review Board of the West China Hospital, Sichuan
produced by a PS under laser irradiation at a specific wavelength to University, Chengdu, China (2020 Review, No. 330). All animal pro­
impair biological activity of tumor cells and ultimately induce cell death cedures were performed in accordance with the Guidelines for Care and
[28,29]. Therefore, PDT has been considered to be an ideal, flexible and Use of Laboratory Animals of West China Hospital, Sichuan University
non-invasive method for precise treatment of tumors. and approved by the Animal Ethics Committee of China. Human tumor
Although many polymeric nanomedicines show encouraging thera­ samples were obtained with written consent from patients. All patients
peutic effects in animal models, their clinical performances are often far had been pathologically diagnosed with muscle invasive bladder cancer
below the initial expectations. Failure to translate animal study results and underwent radical cystectomy at West China Hospital, Sichuan
into clinical outcomes is often associated with the limitations of cell lines University. Their tumor tissues were collected immediately after surgery
and cell line-derived xenografts in preclinical animal studies due to the and minced into pieces and injected to the NSG mouse (NOD SCID
lack of their clinical relevance and heterogeneity [30]. Recently, gamma mouse) via subcutaneous transplantation. When the tumors in
patient-derived tumor xenografts (PDXs) models have been established mice grew to an approximate volume of 500 mm3, they were harvested
to mimic human tumors and they are considered as a more feasible tool and subcutaneously transplanted into BALB/c nude mice. After multiple
for translational research [31]. Importantly, PDXs can maintain the passages (> 3), the PDX tumors could be used for in vivo experiments.
cellular and histopathological structure of their parental tumors, and The tumor volume (mm3) was calculated as (L × W2)/2, where L and W
genomic and proteomic profiles between PDXs and their parental tumors were the length and width of tumors, respectively.
are preserved [32,33]. It is worth noting that the sensitivity to anti­
cancer drugs in PDXs is well correlated with the response of patients to 2.4. PCI-induced internalization
these drugs in clinical treatment [34]. PDXs are a promising model in
predicting the efficacy of conventional and novel anti-cancer therapies. T24 cells were seeded in glass bottom dishes and cultured for 24 h.
Therefore, in this study, we constructed a stimuli-responsive, The medium was replaced with fresh one containing Ce6 or NPs@Ce6 at
amphiphilic, block poly (OEGMA)-PTX prodrug loaded with Ce6, and 0.5 μg/mL of Ce6. Cells were incubated for 1 h and treated with 660 nm
studied its potential as an efficient nanomedicine in the bladder cancer NIR irradiation (0.2 J/cm2). At different incubation durations (2 h, 4 h,
PDX model (Scheme 1). The hydrophilic fragment of the polymer pro­ 6 h), cells were incubated in fresh medium containing 1 × Hoechst
drug was prepared by reversible addition fragmentation chain transfer 33342 for nuclei staining and observed under a confocal laser scanning
(RAFT) polymerization of OEGMA, and an anti-tumor drug PTX was microscope (CLSM). PCI-induced internalization of NPs@Ce6 was
covalently linked to the polymer backbone through a cathepsin B- quantified by ImageJ software. In terms of PCI-induced internalization
responsive tetrapeptide Gly-Phe-Leu-Gly (GFLG) (Scheme S1). The self- in T24 MTS, MTSs were incubated with fresh medium containing 1 ×
assembled nanostructure could efficiently encapsulate Ce6 for PDT and Hoechst 33342 overnight, and then treated with NPs@Ce6 at 0.5 μg/mL
PCI. The therapeutic effect of the multifunctional nanomedicines (poly of Ce6 for 1 h and subjected to 660 nm NIR irradiation (0.5 J/cm2). At
(OEGMA)-PTX prodrug incorporating Ce6) was evaluated in a bladder different incubation time points (2 h, 4 h, 12 h), the MTSs were collected
cancer PDX model, including the examination of the underlying anti­ and analyzed under a microscope.
cancer mechanisms at the genomic and proteomic level. With synergistic
action between the PS and the carrier, the delivery vehicles could 2.5. In vitro antitumor treatment
overcome multiple physiological barriers in the delivery process, and
effectively suppress tumor growth through the combination of chemo­ To evaluate the anticancer effect of the nanoparticles, T24 cells were
therapy and PDT. The anticancer efficacy of the nanomedicine in the seeded in 96-well plates at a density of 3000 cells per well and incubated
bladder cancer PDX model indicates that combinational chemotherapy for 24 h. Cells were treated with different concentrations of free Ce6,
and nanomedicine-facilitated PDT have a great potential in anti-cancer PTX, NPs, and NPs@Ce6 with or without 660 nm NIR irradiation (0.6 J/
treatment. cm2), respectively. For NPs@Ce6+SL group (short-long-term irradia­
tion), cells were subjected to short-term irradiation (0.2 J/cm2) after 1 h
2. Materials and experimental procedures incubation for PCI-induced internalization, and 0.4 J/cm2 irradiation
after 6 h incubation. In the NPs@Ce6+L group (long-term irradiation),
2.1. Materials and methods 0.6 J/cm2 irradiation was given after 6 h incubation. Unless specified
otherwise, the other experiments were performed using the same irra­
The information about materials, measurements and characteriza­ diation protocol. After PDT treatment, cells were further incubated for
tions of poly (OEGMA)-PTX@Ce6 (NPs@Ce6), including preparation, 24 h 10 μL of the cell counting kit-8 (CCK8) reagent was added into each
size characterization, stability, drug release, and additional anticancer well and incubated for 2 h. The absorbance at 450 nm was recorded via a
mechanism studies were provided in Supporting Information. Varioscan Flash microplate reader (Themo Fisher Scientific, USA). The
combination index (CI) of PTX and Ce6 was calculated by CompuSyn
2.2. Tumor cells and tumor spheroids culture software [35].

Human T24 bladder cancer cells were acquired from Shanghai 2.6. Antitumor effect studies in bladder cancer PDX models
Institute for Biological Sciences (China) and cultured by following the
vendor’s instructions. Briefly, cells were maintained in the PRMI 1640 When the tumor volumes reached around 100 mm3, the mice were
medium (Gibco, USA) supplemented with 10% fetal bovine serum divided into seven groups (n = 5 for each group): Ce6+L, PTX, NPs,
(Gibco, USA) and 1% antibiotics (100 units/mL penicillin and 100 mg/ NPs@Ce6, NPs@Ce6+SL, NPs@Ce6+L and saline. The nanoparticles
mL streptomycin) in a 5% CO2, humidified incubator at 37 ◦ C. To culture were injected at an equivalent PTX dose of 7 mg/kg and a Ce6 dose of 5
T24 multicellular tumor spheroids (MTSs), T24 cells were seeded in 96- mg/kg. The tumors in the NPs@Ce6+SL-treated group were subjected to
well U-bottom plates at a density of 2 × 104 cells per well. After 1 week, NIR irradiation (660 nm, 3 J/cm2 = 2 min × 25 mW/cm2) for PCI-
tumor spheroids were formed with a diameter of approximately 200 μm. induced internalization after 1 h injection, and mice in this group
received second NIR irradiation (6 J/cm2 = 4 min × 25 mW/cm2) at 6 h
after injection. The tumors in both Ce6+L and NPs@Ce6+L groups

3
P. Tan et al. Biomaterials 277 (2021) 121061

received NIR irradiation (9 J/cm2) after 6 h injection, while other groups Control, NPs@Ce6+L versus Control, NPs@Ce6+L versus Ce6+L, and
were kept in dark. The above anticancer treatment procedure was per­ NPs@Ce6+L versus NPs were used in Gene Ontolog by clusterProfiler
formed every 4 days and four times in total. The tumor sizes and body (v3.10.1) [40] and gene set enrichment analysis (GSEA) [41,42].
weights were recorded every two to four days. All mice were sacrificed
at day 19, tumors and main organs of each group were harvested. The 2.9. Statistical analysis
tumors in each group were weighed. The tumors and main organs were
treated with Hematoxylin and Eosin (H&E) and immunohistochemistry Student’s t-test was applied for statistical analysis of experimental
(IHC) staining. In addition, fresh tumor tissues in the control, Ce6+L, data in comparison with the controls. All data were presented as mean ±
NPs and NPs@Ce6+L groups were used for bulk RNA-seq to unveil the standard deviation (SD). Statistical significance was indicated by P <
anticancer mechanisms. 0.05 and highly statistical significance by P < 0.01.

2.7. Western blots 3. Results and discussion

T24 Cells were treated in the same way as in the apoptosis assay. The 3.1. Preparation and characterization of NPs@Ce6
treated cells were harvested, and the cell lysate was immediately pre­
pared in a RIPA buffer containing protease inhibitors and phosphatase Ce6-loaded nanoparticles (NPs@Ce6) were prepared by encapsu­
inhibitors. Electrophoresis was performed according to a standard pro­ lating Ce6 with an enzyme-responsive diblock poly (OEGMA)-PTX
tocol using 10% SDS-polyacrylamide gel and transferred to Immobilon-P prodrug. To obtain the prodrug, a cathepsin B-responsive monomer
(Millipore). Primary antibodies were applied at 1:500–1:2000 dilution (MA-GFLG-PTX) was first synthesized according to a previously reported
in 5% BSA in TBST for incubation overnight at 4 ◦ C. Membranes were method [43]. Subsequently, a hydrophilic macromolecular chain
incubated with HRP-conjugated secondary antibodies (1:10000) for 1 h transfer agent, poly (OEGMA)-CTA, was prepared by RAFT polymeri­
at room temperature. The proteins were visualized using NcmECL Ultra zation as sketched in Scheme S1 (Supporting Information). The GPC
Reagent (NCM biotech). The Immobilon® Western Chemiluminescent results showed that the poly (OEGMA)-CTA had a high molecular weight
HRP Substrate (WBKLS0500, MILIPORE) was used for detection of and a narrow molecular weight distribution (Mn = 25.8 kDa, PDI =
proteins. 1.13). Compared with many PEG-modified polymers, this poly (OEG­
MA)-based hydrophilic fragment has a significantly higher molecular
2.8. RNA-seq analysis weight, so it was expected to have a better assembly performance and
encapsulation properties.
2.8.1. Library construction The proportion of hydrophilic and hydrophobic segments in polymer
According to a standardized Illumina protocol, tumor tissues from prodrugs has an important impact on its performance. In this study, the
the PDX mouse model at the end of treatment (day 19) were used for the content of PTX in the polymer can be adjusted by controlling the reac­
preparation and construction of transcriptome libraries. Subsequent tion conditions. To obtain polymeric prodrugs with optimized physico­
transcriptome sequencing analysis was performed on the Illumina chemical properties, we first studied the self-assembly properties of
NovaSeq 6000 platform. polymer prodrugs obtained after copolymerization of the poly
(OEGMA)-CTA and MA-GFLG-PTX monomer through dissipative parti­
2.8.2. Pre-processing raw data cle dynamics (DPD) simulation. As shown in Figure S1 (Supporting In­
After 150 bp paired-end sequencing raw data was obtained, adapter, formation), we constructed the structural models of polymer prodrugs
ploy-N, and low-quality reads were removed to obtain high-quality polymerized with 2–5 MA-GFLG- PTX respectively and studied their self-
clean data. Clean data was checked against the criteria of Q20 > 90 assembly processes and structures through DPD simulation (Figure S2
and Q30 > 85. Clean data was used for subsequent analysis of alignment, and S3, Supporting Information). The DPD simulation results showed
differential genes, and pathway enrichment. that the poly (OEGMA)-PTX prodrug with different contents of PTX (the
theoretical range is 6.0%–13.2 wt%) could self-assemble to form nano­
2.8.3. Sequence alignment and differential expression analysis particles with a hydrophobic inner core constituted with PTX and GFLG,
Sequence alignment was executed with STAR (v 2.6.0) [36] using the and a hydrophilic shell from poly (OEGMA) (Fig. 1a). However, as the
reference human genome (GRCh38). Deseq2() [37] was performed to PTX drug content in the prodrug increased, the outer poly (OEGMA)
generate normalized expression data with the option of “counts segment could not completely cover the inner PTX molecules, and the
(normalized = TRUE)”. The significance score and log2fold-change were morphology of the formed nanoparticles gradually changed from a
calculated. complete core-shell structure to a semi-packaged structure. Although a
low content of PTX was more conducive to self-assembly of the prodrug,
2.8.4. Drawing heatmap and volcano map such an amount of PTX resulted in a lower drug loading into the drug
The heat map matrix was composed of different genes with a p-value delivery system. To balance a high drug content for a desired therapeutic
< 0.05 by comparing genes from the treatment group and those from the index and successful self-assembly for a complete core-shell structure,
control group, and the data normalization was performed by DESeq2. the polymer structure containing three GFLG-PTX repeating units may
The function adapted from Seurat [38,39] DoHeatmap() was used for be an ideal polymeric prodrug structure. Therefore, we used the results
visualization and gene labeling. The volcano map matrix included of DPD simulation as a guide and optimized the reaction conditions to
log2Foldchange and -log10*(p_value) which were obtained by copolymerize MA-GFLG-PTX and poly (OEGMA)-CTA to prepare an
comparing NPs@Ce6+L versus Ce6+L by DESeq2. Red plots were drawn amphiphilic block polymer prodrug poly (OEGMA)-PTX.
for genes up-regulated in the NPs@Ce6+L-treated group (p-value < 0.05 Compared with the 1H NMR spectrum of poly (OEGMA)-CTA
and log2foldchange > 0.3), and blue plots for genes down-regulated in (Figure S4, Supporting Information), a characteristic peak of PTX
the NPs@Ce6+L-treated group (p_value < 0.05 and log2foldchange < (7.25–8.00 ppm) was observed in the 1H NMR spectrum of poly
-0.3). Data visualization and gene labeling were performed by ggplot2 (OEGMA)-PTX (Figure S5, Supporting Information), which indicated
(https://ggplot2.tidyverse.org) and ggrepel. that PTX was successfully conjugated to the polymer through a tetra­
peptide GFLG. The results of amino acid analysis showed that the
2.8.5. Pathway enrichment analysis percent weight of the amino Gly, Phe, and Leu were 1.19%, 1.42%, and
Differentially expressed genes (p_value < 0.05 and |log2fc| > 0.3) 1.21%, respectively, further indicating the presence of GFLG in the
which were identified by comparing samples between NPs versus polymer (Table S1, Supporting Information). GPC confirmed a

4
P. Tan et al. Biomaterials 277 (2021) 121061

Fig. 1. Characterization of NPs and NPs@Ce6. a) Structures of nanoparticles formed by polymer prodrugs with different PTX drug contents obtained by DPD
simulation. b) Determination of CMC of the poly (OEGMA)-PTX prodrug using pyrene as a fluorescence probe. Hydrodynamic size distributions and TEM images of c)
NPs and d) NPs@Ce6. e) Colloidal stability of NPs@Ce6 in different solutions. f) Photostability of NPs@Ce6 and free Ce6 dissolved in PBS or DMSO. g) Time-
dependent recovery of SOSG fluorescence from free Ce6, NPs@Ce6 or NPs under light exposure (660 nm, 2 mW/cm2). h) HPLC chromatograms of poly
(OEGMA)-PTX incubated with papain for 0 h (middle) and 12 h (down), with free PTX used as a control (up). i) In vitro PTX release profiles from NPs@Ce6 in
different conditions. j) In vitro Ce6 release profiles from NPs@Ce6 in different conditions.

molecular weight of 28.3 kDa for poly (OEGMA)-PTX and a PDI of 1.28. could form stable micelles while encapsulating hydrophobic therapeutic
The cleavage of GFLG by papain that has a similar activity to cathepsin B agents.
could specifically release covalently linked PTX. The content of PTX in To obtain a better anti-tumor effect, a Ce6-encapsulated polymer
poly (OEGMA)-PTX was determined to be 8.3% by HPLC analysis, nano drug delivery system (NPs@Ce6) was prepared by a thin-film hy­
indicating that on average, each polymer prodrug contained 2.7 PTX dration method, so that it could be used for chemical-photodynamic
repeating units. synergistic therapy. As shown in Fig. 1d, the prepared NPs@Ce6 has a
The self-assembly structure was further examined from its 1H NMR uniform spherical structure, and the hydrodynamic particle size was
spectrum. As shown in Figure S6 (Supporting Information), all peaks of about 168.20 ± 1.12 nm. In addition, NPs@Ce6 showed excellent
the polymer prodrug were observed in (CD3)2SO, while only the peak for colloidal stability in phosphate buffered saline (PBS) containing 10%
the hydrophilic OEGMA chain was seen in D2O, indicating that it had a fetal bovine serum, and cell culture media (Fig. 1e). The results of
stable core-shell structure in an aqueous solution. The critical micelle photostability (Fig. 1f) and ROS generation (Fig. 1g) confirmed that the
concentration (CMC) value of poly (OEGMA)-PTX was measured to be encapsulated Ce6 had similar photostability/photoactivity to free Ce6,
20.0 μg/mL using pyrene as a fluorescence probe (Fig. 1b). The hydro­ and it could rapidly produce cytotoxic ROS under 660 nm laser irradi­
dynamic diameter of the nanoparticles formed by the polymer prodrug ation, indicating NPs@Ce6 could exert their PDT effect against tumors.
was 131.96 ± 8.16 nm by DLS (Fig. 1c), which was close to the result of To establish the release profiles of PTX and Ce6 triggered by
TEM. These results indicated that the amphiphilic polymer prodrug cathepsin B in tumor cells, NPs@Ce6 were incubated in a simulated

5
P. Tan et al. Biomaterials 277 (2021) 121061

tumor microenvironment (McIIvaine’s buffer at pH 5.4 with papain for 3.2. Uptake of NPs by cells and tumor spheroids
mimicking cathepsin B) in comparison to a normal physiological envi­
ronment (McIIvaine’s buffer without papain at pH 7.4). HPLC detection The uptake of free Ce6 and NPs@Ce6 by T24 cells was studied via
results showed that the absorption peak for free PTX was detected after flow cytometry and CLSM. As shown in Figure S8 (Supporting Infor­
the NPs@Ce6 were incubated with papain (Fig. 1h and Figure S7, Sup­ mation), the fluorescence intensity of Ce6 (red) in the cytoplasm was
porting Information), indicating that the polymer prodrug could release very weak in T24 cells post incubation with free Ce6 from 2 to 6 h. The
the original PTX drug through enzymatic cleavage. As shown in Fig. 1i, fluorescence intensity of Ce6 was significantly higher in T24 cells after
PTX was released rapidly in McIIvaine’s buffer containing papain at pH they were treated with NPs@Ce6 compared to those treated with free
5.4, and the release amount reached 80% within 8 h. In contrast, in PBS Ce6. As shown in Fig. 2a and b, the cellular fluorescence intensity of Ce6
(pH 7.4 or 5.4) without papain, the release of PTX was less than 20% in gradually increased in T24 cells treated with NPs@Ce6, suggesting that
24 h. The release profile of Ce6 in different buffer solutions was similar the internalization of NPs@Ce6 by T24 cells was time dependent. A
to PTX, and rapid release of Ce6 from NPs@Ce6 occurred in the presence relatively low dose of light (0.2 J/cm2) was used to evaluate the PCI
of papain (Fig. 1j). These results indicated that NPs@Ce6 were respon­ effect on the cellular uptake of NPs@Ce6. A notably enhanced fluores­
sive to cathepsin B, and responsive degradation of GFLG unpacked the cence signal of Ce6 was observed at 2 h, 4 h, and 6 h in cells exposed to a
core-shell nanostructure, thereby quickly releasing the contained drugs. low dose of NIR light irradiation (Fig. 2a). Semi-quantitative analysis
confirmed that NIR light irradiation induced statistically significant
cellular uptake of NPs@Ce6 (Fig. 2b).

Fig. 2. PCI-induced internalization and penetration. a) CLSM images for intracellular localization of NPs@Ce6 (0.5 μg/mL of Ce6) at different time-points with or
without NIR light irradiation at 0.2 J/cm2. Hoechst 33342 (blue) for nuclei, and fluorescence of Ce6 (red) for NPs@Ce6. Scale bar: 20 μm. b) Semi-quantitative
analysis of integrated fluorescence density of intracellular Ce6 of Fig. 2a. *P < 0.05, ***P < 0.001, ****P < 0.0001 (n = 16). c) CLSM images of T24 spheroids
after treatment with NPs@Ce6 (0.5 μg/mL of Ce6) for 12 h with or without NIR light irradiation at 0.5 J/cm2. Purple for the fluorescence of Ce6. Scale bar: 100 μm.
d) Fluorescence distribution of Ce6 in T24 spheroids along the cross-section of the spheroid marked in Fig. 2c e) Flow cytometry histograms and f) mean fluorescence
intensity (MFI) of Ce6 in T24 cells after treatment with NPs@Ce6 for different durations with or without NIR light irradiation at 0.2 J/cm2 ***P < 0.001 (n = 3). (For
interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

6
P. Tan et al. Biomaterials 277 (2021) 121061

As the incubation time increased, the fluorescence intensity in the the PCI effect.
spheroids significantly increased, indicating time-dependent penetra­ Microtubules, one of the intracellular cytoskeletal filaments, play a
tion of NPs@Ce6 (Figure S9, Supporting Information). In addition, the critical role in mitosis and intracellular vesicular trafficking. PTX has
fluorescence intensity in T24 spheroids at the incubation time ranging been demonstrated to be able to stabilize microtubules and suppress
from 2 to 12 h was enhanced after the spheroids received NIR light their dynamicity, thereby blocking mitosis and inducing cell death. To
irradiation (0.5 J/cm2), indicating the penetration of NPs@Ce6 was evaluate the impact of PTX from NPs@Ce6 on the microtubules of T24
enhanced by the PCI effect (Fig. 2c and d and Figure S9, Supporting cells, the morphology of microtubules in cells after treatment with
Information). different formulations was observed under a CLSM. As shown in Fig. 3d,
Flow cytometry analysis also revealed that free Ce6 was not readily the microtubules congregated around the cellular cortex in cells treated
uptaken by T24 cells and supported time-dependent internalization of with PTX, NPs and NPs@Ce6, while microtubules were evenly distrib­
NPs@Ce6 by T24 cells (Fig. 2e and f). Importantly, flow cytometry uted throughout the cytoplasm in the control and Ce6-treated groups. In
analysis also demonstrated that the low dose irradiation increased the addition, the cells in the control and Ce6-treated groups maintained an
uptake of nanoparticles through the PCI effect. Thus, these results sug­ integrated and intact nuclear structure, while multiple disintegrated
gested that photodynamic treatment at a low dose could promote nuclei structures were observed in the cells treated with PTX, NPs and
cellular uptake of NPs@Ce6 and enhance their spheroid penetration. In NPs@Ce6, indicating PTX released from NPs and NPs@Ce6 exerted a
addition, live cell images revealed red fluorescence of NPs@Ce6 was similar effect on disruption of mitosis and damage of DNA to free PTX.
largely observed in regions that was stained with Lysotracker green, Furthermore, the impact of PTX released from NPs@Ce6 on cell cycle
implying intracellular distribution of NPs@Ce6 was in the lysosomes of progression of T24 cells was assessed by PI staining and detected by flow
T24 cells (Figure S10, Supporting Information). cytometry. As shown in Fig. 3e and f, the cells in the control and Ce6+L
groups were mainly in the G0/G1 phase (~66%) and the S phase
3.3. Imaging of ROS generation (~25%), while only ~7% cells were in the G2/M phase. After T24 cells
were treated with PTX, NPs, or NPs@Ce6, the proportion of cells in the
The intracellular ROS level after PDT treatment was measured via G2/M phase significantly increased to 85.83%, 83.45%, and 83.52%,
DCFH-DA in T24 cells and spheroids. As shown in Fig. 3a and Figure S11 respectively; and the percentages of cells in the G0/G1 and S phase were
(Supporting Information), the ROS level significantly increased in cells significantly decreased to around 1% and 15%, respectively. These re­
treated with NPs@Ce6 and PDT treatment compared with those that sults suggested that PTX released from NPs or NPs@Ce6 could induce
received the free Ce6 and PDT treatment (abbreviated as Ce6+L). cell cycle arrest in the G2/M phase of mitosis in a similar manner as free
Moreover, the production of ROS in cells or tumor spheroids was PTX, which led to cancer cell death.
enhanced in the NPs@Ce6+SL group in comparison with the Apoptosis of T24 cells induced by the combinational PTX and PDT
NPs@Ce6+L group, which may be contributed to the PCI effect induced treatment was also evaluated by flow cytometry (Fig. 3g and h). The
by low-dose light irradiation. The cellular ROS generation was also results revealed that the treatment with Ce6 and NIR irradiation did not
noticed to be enhanced after cells were treated by PTX, NPs, NPs@Ce6, increase cell apoptosis, which may be due to extremely poor cellular
which was consistent with previous reports that PTX could induce ROS uptake of free Ce6 by T24 cells. However, exposure to free PTX and PTX
generation [44,45]. from NPs or NPs@Ce6 could significantly induce apoptosis, and
apoptotic cells accounted for 27.87% of the PTX-treated group, 38.37%
3.4. In vitro anticancer treatment of the NPs-treated group and 49.37% of the NPs@Ce6-treated group.
Moreover, a combinational PDT and PTX treatment further increased
The synergistic antitumor effect of PTX and PDT was evaluated by the percentage of apoptotic cells; and more apoptotic cells were found in
staining live and dead cells with Calcein AM and PI in T24 spheroids. the NPs@Ce6+SL group compared with the NPs@Ce6+L group (83.4%
The fluorescence images confirmed that treatment with NPs@Ce6 plus vs 70.17%, p < 0.0001), which may be attributed to the PCI effect due to
PDT induced more dead cells (red) in comparison with chemotherapy or photo-responses of Ce6 under NIR illustration to enhance penetration of
PDT treatment alone (Fig. 3b). In addition, more dead cells were NPs@Ce6 in tumor spheroids and cellular uptake of NPs@Ce6 to
observed in the NPs@Ce6+SL group than those in the NPs@Ce6+L improve their toxicity to tumor cells.
group. Moreover, the CCK8 was employed to evaluate the cytotoxicity
against T24 cells induced by NPs@Ce6 plus PDT treatment. As shown in 3.5. In vivo anticancer treatment
Fig. 3c, treatment with free Ce6 did not impact the viability of T24 cells
even when the concentration of free Ce6 was increased up to around 7 To assess the biodistribution of NPs@Ce6 after systemic adminis­
μg/mL, but the cell viability decreased when the cells were treated with tration and their in vivo therapeutic efficacy, a human bladder cancer
free Ce6 at a concentration of more than 1 μg/mL and subjected to NIR PDX model was established for imaging and anticancer treatment.
light irradiation (0.6 J/cm2). In addition, there was no statistically sig­ Briefly, the bladder cancer tissue was extracted from a patient who was
nificant difference in the cell viability between the Ce6+SL and Ce6+L diagnosed with muscle invasive bladder cancer in the West China Hos­
group, indicating free Ce6 under NIR irradiation was unable to exert a pital and underwent radical cystectomy. H&E staining of the cancer
PCI effect on cells. Along with an increase in the PTX concentration, the tissue confirmed the diagnosis of bladder cancer in this patient (Fig. 4a).
cell viability gradually decreased after treatment with PTX, NPs or To establish a bladder cancer PDX model, the fresh tumor tissue was
NPs@Ce6. The IC50 (PTX) values of NPs (11.2 μg/mL) and NPs@Ce6 minced into pieces at ~1 mm3 and subcutaneously implanted into NSG
(5.7 μg/mL) were higher than that of PTX (3.5 μg/mL), which may be mice. H&E staining of the xenografts from mice after several in vivo
attributed to a more complex process of internalization, trafficking, and passages showed that the PDX model inherited the histopathological
drug release of NPs and NPs@Ce6 than free PTX. Notably, the IC50 (PTX, features of the parental tumor. Thus, the established PDX model
Ce6) value of NPs@Ce6+SL (0.16 μg/mL) was lower than that of mimicking the human bladder cancer could be used for evaluating the
NPs@Ce6+L (0.38 μg/mL). In addition, the CI values of PTX and Ce6 therapeutic effect of drugs in vivo. The bladder cancer PDX model at the
were less than one, indicating a synergistic antitumor effect between fifth passage was employed for assessment of in vivo biodistribution and
them (Figure S12, Supporting Information). Treatment with PTX, NPs, therapeutic outcomes of NPs@Ce6.
NPs@Ce6 with or without NIR light irradiation also significantly
inhibited the cloning formation of T24 cells (Figure S13, Supporting 3.5.1. Pharmacokinetics and biodistribution
Information). These results may be attributed to a better penetration Ce6 is a fluorescence probe, which exhibits deep tissue penetration
ability and a higher cellular uptake efficiency of NPs@Ce6 induced by and low autofluorescence [46]. Thus, through detection of Ce6

7
P. Tan et al. Biomaterials 277 (2021) 121061

Fig. 3. In vitro anticancer treatment. a) ROS generation in T24 tumor spheroids treated with Ce6, PTX, NPs and NPs@Ce6 with or without NIR light irradiation.
NPs@Ce6+SL (0.5 J/cm2 for PCI-induced internalization at 1 h of incubation, and 1.0 J/cm2 at 6 h of incubation). Scale bar: 100 μm. b) CLSM images of T24 tumor
spheroids after treatment with Ce6, PTX, NPs and NPs@Ce6 with or without NIR light irradiation. Live cells were stained with Calcein AM (green) and dead cells with
PI (red). Scale bar: 100 μm. c) Cytotoxicity of Ce6, PTX, NPs and NPs@Ce6 with or without NIR light irradiation towards T24 cells (n = 3). NPs@Ce6+SL (0.2 J/cm2
for PCI-induced internalization at 1 h of incubation, and 0.4 J/cm2 at 6 h of incubation). d) Microtubule aggregation in T24 cells after treatment with Ce6, PTX, NPs
and NPs@Ce6 with or without NIR light irradiation for 12 h. Red fluorescence for tubulin with Tubulin-Tracker Red and blue for nuclear with DAPI. Scale bar: 20 μm.
e, f) flow cytometry analysis of cell cycle distribution in T24 cells after treatment with Ce6, PTX, NPs and NPs@Ce6. Mean ± SD (n = 3). *P < 0.05, ****P < 0.0001.
g, h) flow cytometry analysis of apoptosis in T24 cells after treatment with Ce6, PTX, NPs and NPs@Ce6 with or without NIR light irradiation. Mean ± SD (n = 3),
****P < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

8
P. Tan et al. Biomaterials 277 (2021) 121061

Fig. 4. In vivo anticancer treatment. a) Establishment of a human bladder cancer PDX model in BALB/C nude mice for evaluation of the antitumor effect with
representative H&E staining images for the parental tumor and xenografts of the first (F1), third (F3) and fifth generation (F5). Xenografts (F5) were used for in vivo
imaging and drug treatment. b) In vivo fluorescent images of mice bearing PDX tumors treated with NPs@Ce6 and free Ce6 (5 μg/mL) at different time-points, and ex
vivo fluorescent images for harvested tumors and major organs at the end of in vivo imaging (48 h). c) Representative CLSM images of tumor tissues treated with
NPs@Ce6 or Ce6 for 12 h. Red fluorescence denotes to Ce6, Green fluorescence denotes to CD31 for blood vessels, and DAPI fluorescence denotes to nuclei. Scale bar:
100 μm. d) In vivo pharmacokinetic curves of the Ce6 concentration after intravenous injection of NPs and Ce6, respectively (Mean ± SD, n = 5). e) Experiment
procedure for injection of different formulations with 4 doses and light irradiation. f) Representative whole-animal photographs of mice bearing PDX tumors after
treatment. Red circles refer to the region with PDX tumors. g) Changes in tumor volume in mice treated with saline, Ce6, PTX, NPs and NPs@Ce6 with or without NIR
light irradiation (660 nm, 6 min at 250 mW/cm2), *P < 0.05, **P < 0.01 and ns = not significant (n = 5). PCI-induced internalization and penetration were
performed under 660 nm NIR light irradiation for 2 min at 250 mW/cm2 h) Tumor weights from bladder cancer PDX mice after treatment (Mean ± SD, n = 5, **P <
0.01). i) Photographs of tumors harvested from mice on day 19 after different treatments. Red circles refer to eliminated tumors. j) Images of H&E staining, IHC
staining for Cleaved Caspase-3, and TUNEL assays for bladder cancer PDX tumors. In the IHC staining, brown for apoptotic cells. In TUNEL assays, green for apoptotic
cells. Scale bar: 100 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

fluorescence, it was feasible to reveal dynamic distribution of NPs@Ce6 enhanced accumulation of NPs@Ce6 at the tumor site was observed
in a specific organ or tissue and their excretion pathways. In vivo and ex with an increase in the post-injection time compared with free Ce6,
vivo bioimaging was performed in PDX mice after intravenous injection indicating NPs@Ce6 could significantly increase Ce6 accumulation at
of NPs@Ce6 or free Ce6. Quantitative fluorescent signals of Ce6 from tumor sites. After in vivo imaging at 48 h, main organs and tumors of
NPs@Ce6 and free Ce6 at different time points were shown in Fig. 4b mice were harvested and imaged. The fluorescent signal was strongest in
and Figure S14 (Supporting Information), respectively. Noticeably tumors compared with other organs. In addition, the tumor tissues were

9
P. Tan et al. Biomaterials 277 (2021) 121061

harvested at 12 h after injection and imaged under a CLSM. The images with the NPs@Ce6+L group, which may be due to the PCI effect
revealed the fluorescence intensity of Ce6 in tumors treated with (Fig. 4j). In addition, Ki-67 staining showed that proliferation of tumor
NPs@Ce6 was more intense compared to that in tumors treated with free cells was significantly inhibited after treatment by NPs@Ce6+SL and
Ce6 (Fig. 4c). Moreover, in vivo pharmacokinetic analysis showed the NPs@Ce6+L (Figure S19, Supporting Information). These results
concentration of Ce6 in the blood of mice treated with NPs@Ce6 demonstrated that the combinational treatment of chemotherapy and
reduced at a relatively slower rate, with a half-life of 447.5 min, than PDT effectively inhibited proliferation of tumor cells and induced their
that in the mice treated with free Ce6 with a half-life of 79.1 min (Fig. 4d apoptosis.
and Table S2, Supporting Information). This may be attributed to a high To further unveil the underlying molecular mechanisms of the
MW of NPs@Ce6, a stable linker in the chemical structure, a slightly anticancer effect from the combinational approach with NPs@Ce6
negatively charged surface, and a branched structure with excellent under NIR illustration, bulk RNA-sequencing was performed to compare
chain flexibility and deformability. The prolonged circulation time the transcriptomics of tumor samples extracted from mice after different
could contribute to the passive accumulation of NPs@Ce6 at the tumor treatments. The heatmap showed differentially expressed genes (DEGs)
site and boost their therapeutic effects. in tumors subjected to the combinational treatment with PTX and PDT
(Fig. 5a). The DEGs included SOD2, BID, H2AX, COA3, NDUFAB2, PIM1,
3.5.2. Antitumor therapy and TACC3. Among them, TACC3 was involved with spindle, H2AX with
The in vivo synergistic anticancer efficacy was evaluated in bladder DNA damage, and SOD2, COA3 and NDUFAB2 with reactive oxygen
cancer PDX models. Mice bearing bladder tumors were divided into 7 species (ROS) generation pathways. The volcano plots showed the on­
groups (n = 5 for each group). The control group was intravenously cogenes, such as KLF9, STC1, IGF2, ABCC1 and CAPN6, were down-
injected with saline. For PCI-induced internalization, the tumors first regulated, while the tumor suppressor genes including CDH11 and
received NIR light irradiation at 20 mW/cm2 for 2 min at 1 h after in­ SMOC2 were up-regulated in the NPs@Ce6+L group compared with the
jection of NPs@Ce6. A total of 4 doses were administrated to mice Ce6+L group (Fig. 5b). Gene Ontolog (GO) term enrichment analysis
during the entire treatment procedure. Changes in the tumor volume revealed that up-regulated genes after treatment with NPs@Ce6+L
were monitored every 2–4 d and mice were photographed after treat­ compared with Ce6+L were mainly involved in ROS generation, mito­
ment on day 19 (Fig. 4e and f). During the 19-day treatment, the tumors chondrial respiratory pathways, DNA damage and mitosis-associated
rapidly grew with a 13.8-fold increase in the volume in the mice that pathways, including oxidative phosphorylation, activation of the mito­
received saline, while PDT treatment with free Ce6 slightly inhibited chondrial respiratory chain, impedance of the repair function for broken
tumor growth with a 9.8-fold increase in the tumor volume, which may double-strand DNA and prevention of disassembly of spindle microtu­
be due to the relatively weak enhanced permeability and retention bules (Fig. 5c). Meanwhile, down-regulated genes associated with
(EPR) effect and poor cellular uptake of free Ce6 by tumor cells mitosis-related pathways were significantly enriched in the NPs group
(Fig. 4g–i), which were consistent with the in vitro effect of free Ce6. In compared with the control group (Fig. 5d).
addition, free PTX, NPs and NPs@Ce6 without PDT treatment showed a Gene set enrichment analysis (GSEA) indicated that the gene sets
similar inhibition efficacy with a 6-fold, 4.8-fold and 5.3-fold increase in associated with oxidative phosphorylation (OXPHOS, NES = 1.75, FDR
the tumor size, respectively. However, the combinational treatment = 0.00), DNA repair (NES = 1.38, FDR = 4.84E-3) and mitochondria
(NPs@Ce6+L and NPs@Ce6+SL) displayed a significantly improved gene module (NES = 1.83, FDR = 0.00) had significantly positive
inhibitory effect on tumor growth compared with chemotherapy via free enrichment scores in the tumors treated with NPs@Ce6+L compared
PTX, NPs and NPs@Ce6. Notably, the tumors in the NPs@Ce6+SL with those with Ce6+L treatment, while gene sets associated with TGF-β
treatment group were significantly smaller with a 5-fold reduction in the signaling (NES = − 1.39, FDR = 0.03), hypoxia_UP (NES = − 1.73, FDR
original tumor volume compared with those that were subjected to = 0.00) and TNF-α signaling via NF-κB (NES = − 1.46, FDR = 0.03) had
NPs@Ce6+L treatment with a 3-fold increase in the tumor volume. The significantly negative enrichment scores (Fig. 5e), which confirmed the
inhibition score of tumors in the NPs@Ce6+SL group reached 98.5%. results from GO term analysis. Meanwhile, gene sets related to prolif­
These results were consistent with tumor growth kinetics (Figure S15, eration, division, cell cycle and mitosis were negatively enriched in tu­
Supporting Information), average tumor weights (Fig. 4h) and mor­ mors treated with NPs, such as Myc target_V1, cell division, G2M
phologies of tumors upon harvest on day 19 (Fig. 4i). The tumors in two checkpoint and proliferation, compared with those treated with Saline
mice in the NPs@Ce6+SL treatment group were completely eradicated (Figure S20, Supporting Information). When gene sets were compared
(Fig. 4i), which may be ascribed to passive accumulation and PCI- between tumors treated with NPs and those with NPs@Ce6+L, gene sets
induced penetration of NPs@Ce6 at the tumor sites and PCI-enhanced for oxidative phosphorylation (NES = 1.56, FDR = 0.008) had a positive
cellular uptake of NPs@Ce6 to exert a synergistic therapeutic effect enrichment score, while TGF-β signaling (NES = − 1.66, FDR = 0.003)
after rapid release of PTX and Ce6 under NIR irradiation. Moreover, and TNF-α signaling via NF-κB (NES = − 1.25, FDR = 0.19) had negative
slight fluctuations in the body weight were observed in mice from all enrichment scores (Figure S20, Supporting Information).
groups, but there was no significant difference in the body weight Mitochondrial ROS are generated mainly on the inner mitochondrial
change among all groups (Figure S16, Supporting Information). In membrane during the process of oxidative phosphorylation. The level of
addition, no obvious pathological changes were seen in major organs oxidative phosphorylation is universally downregulated in cancer due to
after treatment by NPs@Ce6+SL, NPs@Ce6+L, NPs@Ce6, NPs, PTX, glycolysis [47]. Down-regulated OXPHOS is often associated with poor
Ce6+L, or saline on day 19, which indicated that NPs@Ce6 had great clinical outcomes across all cancer types and closely correlated with
biocompatibility (Figure S17, Supporting Information). invasion and metastasis of tumors [48,49]. However, a high level of
mitochondrial ROS generated by the upregulated OXPHOS pathway
3.5.3. Mechanisms of the combined PDT and PTX treatment could activate apoptosis pathways that is capable of inducing cell death
All mice were euthanized after the treatment and the tumors were [50]. Our data supported that OXPHOS was up-regulated by the
extracted for mechanistic analysis. H&E staining of tumor tissues NPs@Ce6+L treatment for production of mitochondrial ROS to induce
revealed that there was a large area of necrosis and fragmented nuclei cell apoptosis.
were observed in the NPS@Ce6+SL-treated group (Fig. 4j and It has been suggested that TGF-β could promote cancer progression
Figure S18, Supporting Information). IHC staining of cleaved caspase-3 by increasing invasion and metastasis of tumor cells, and the TGF-β
and the TUNEL assay in tumor tissues indicated the combinational activity is negatively correlated with clinical outcomes [51]. Mean­
treatment with PTX and PDT significantly induced more apoptotic cells while, TNF-α could also promote tumor invasion and metastasis via the
(Fig. 4j and Figure S18, Supporting Information). Moreover, more NF-κB pathway [52]. In particular, it has been revealed that hypoxia in
apoptotic cells were produced in the NPs@Ce6+SL group compared tumors contributes to tumor progression, angiogenesis, metastasis and

10
P. Tan et al. Biomaterials 277 (2021) 121061

Fig. 5. Mechanisms for combining PDT with PTX for anticancer treatment. a) Heatmap of differentially expressed genes (DEGs) in tumors from bladder cancer PDX
models after treatment, measured by RNA-seq. Red for upregulation and blue for downregulation. P adjusted value < 0.05. Three biologically independent animals
per group (n = 3). Key DEGs were listed on the right. b) Volcano plots of DEGs in tumors after NPs@Ce6+L and Ce6+L treatment. c) Gene Ontolog (GO) term
enrichment analysis of up-regulated DEGs after NPs@Ce6+L treatment compared with Ce6+L treatment. d) GO term enrichment analysis of down-regulated DEGs
after NPs treatment compared with the Control; e) Gene signatures with positive enrichment scores from oxidative phosphorylation, DNA repair, and mitochondria
gene module and with negative enrichment scores for TGF-beta signaling, TNFα signaling via NF-κB, and hypoxia from Gene Set Enrichment Analysis (GSEA) in
tumors treated NPs@Ce6+L, comparing to those treated with Ce6+L. f) Western blotting plots of proteins for apoptosis (PARP, cleaved PARP, Caspase-3, Cleaved
Caspase-3, Bcl-2, and Bax), DNA damage (γH2A.X), and cytoskeletal filaments (α/β-tubulin and pan-actin) in T24 cells after treatment with blank, Ce6, PTX, NPs,
NPs@Ce6 with or without NIR light irradiation (660 nm, 0.6 J/cm2). g) Semiquantitative analysis of protein levels in different mice groups from Fig. 5f. ****P <
0.0001. h) Schematic mechanisms of the anticancer effect of the combinational PTX and PDT treatment against bladder cancer. (For interpretation of the references
to colour in this figure legend, the reader is referred to the Web version of this article.)

resistance to therapy [53]. The NPs@Ce6+L treatment resulted in caspase-3 compared with those treated with PTX, NPs or NPs@Ce6
down-regulation of genes related to TGF-β, TNF-α and hypoxia for pre­ (Fig. 5f and g). However, cells treated with Ce6 and Ce6+L did not show
venting tumor progression and metastasis. any difference in the expression of apoptotic proteins compared with
Besides bioinformatics analysis, proteins related to apoptosis, DNA those without any treatment, which may be due to the poor cellular
damage, and cytoskeleton aggregation in bladder cancer T24 cells were uptake of Ce6 by bladder cancer cells. In addition, protein expression of
also analyzed at the proteomics level (Fig. 5f–g, and Figure S21, Sup­ the DNA damage-associated gene γH2A.X was significantly decreased in
porting Information). The protein expression analysis suggested that cells after PTX treatment alone or combined PTX and PDT treatment
apoptosis-related genes including cleaved caspase-3, cleaved PARP, and (Fig. 5f and g). Interestingly, expression of microfilament and
Bax were upregulated after treatment with free PTX, NPs@Ce6, or microtubules-associated proteins, α/β-tubulin and pan-actin, were
NPs@Ce6+L, while the apoptosis-inhibiting gene Bcl-2 was down- reduced after treatment with PTX, but PDT treatment did not alter the
regulated, which suggested the number of apoptotic cells significantly expression of these proteins.
increased in these groups (Fig. 5f and g). Meanwhile, cells treated with In summary, we have revealed the bladder tumor-killing mecha­
NPs@Ce6+L and NPs@Ce6+SL had a higher expression level of cleaved nisms of the combined PTX and PDT treatment via NPs@Ce6 at both

11
P. Tan et al. Biomaterials 277 (2021) 121061

cellular and molecular levels (Fig. 5h). Our results suggested that Credit author statement
chemotherapy by PTX from NPs@Ce6 could block cell division and
proliferation, which resulted in cancer cells death; while a combination Ping Tan: Methodology, Investigation, Experiments, Data curation,
treatment with PTX and PDT via NPs@Ce6 could significantly inhibit Writing – original draft preparation. Hao Cai: Methodology, Investiga­
the genes and pathways related to tumor progression, invasion, and tion, Experiments, Data curation, Writing – original draft preparation.
metastasis, and up-regulate OXPHOS for ROS generation to induce DNA Xiaodi Tang and Yong Yi: Data curation, Formal analysis. Qiang Wei:
damage and cell death. Collected patients’ tissues. Qianfeng Zhang and Junxiao Yang: Data
curation. Michal Kopytynski, Hu Zhang, Qiyong Gong, Rongjun
4. Conclusions Chen and Zhongwei Gu: Reviewed & Edited manuscript. Kui Luo:
Conceptualization, Supervision, Resources, Funding acquisition, Project
We established bladder cancer PDX models for evaluating the ther­ administration, Writing – review & editing.
apeutic efficacy of an enzyme-responsive polymeric nano-drug delivery
system (NPs@Ce6) that efficiently delivered PTX and Ce6 into tumor References
cells. The drug delivery system had great biocompatibility with an
extended circulation time for passive accumulation at tumor tissues. The [1] Y. Sato, T. Nakamura, Y. Yamada, H. Harashima, The nanomedicine rush: new
strategies for unmet medical needs based on innovative nano DDS, J. Contr.
PCI effect induced by short-term irradiation contributed to cellular up­
Release 330 (2021) 305–316.
take and tissue penetration of NPs@Ce6. In the bladder cancer PDX [2] S. Hassan, G. Prakash, A. Bal Ozturk, S. Saghazadeh, M. Farhan Sohail, J. Seo,
models, treatment with NPs@Ce6 with short-term and long-term irra­ M. Remzi Dokmeci, Y.S. Zhang, A. Khademhosseini, Evolution and clinical
translation of drug delivery nanomaterials, Nano Today 15 (2017) 91–106.
diation had the best performance in inhibiting bladder tumors. Bioin­
[3] Q. Zhou, S. Shao, J. Wang, C. Xu, J. Xiang, Y. Piao, Z. Zhou, Q. Yu, J. Tang, X. Liu,
formatics analysis suggested that the combination of PTX chemotherapy Z. Gan, R. Mo, Z. Gu, Y. Shen, Enzyme-activatable polymer-drug conjugate
and PDT via NPs@Ce6 could down-regulate the hypoxia pathway, TGF-β augments tumour penetration and treatment efficacy, Nat. Nanotechnol. 14 (8)
and TNF-α signaling pathways which were involved in tumor progres­ (2019) 799–809.
[4] J. Ding, J. Chen, L. Gao, Z. Jiang, Y. Zhang, M. Li, Q. Xiao, S.S. Lee, X. Chen,
sion, invasion and metastasis, and could arrest the cell cycle, suppress Engineered nanomedicines with enhanced tumor penetration, Nano Today 29
proliferation, and promote oxidative phosphorylation and ROS genera­ (2019) 100800.
tion. Meanwhile, western blots analysis revealed that proteins related to [5] S. Li, Y. Zhang, S.-H. Ho, B. Li, M. Wang, X. Deng, N. Yang, G. Liu, Z. Lu, J. Xu,
Q. Shi, J.-Y. Han, L. Zhang, Y. Wu, Y. Zhao, G. Nie, Combination of tumour-
promoting apoptosis (Bax, cleaved caspase-3, cleaved PARP) and DNA infarction therapy and chemotherapy via the co-delivery of doxorubicin and
damage (γH2A.X) were up-regulated, and those related to inhibiting thrombin encapsulated in tumour-targeted nanoparticles, Nature Biomedical
apoptosis (Bcl-2) and mitosis (pan-actin and α/β-tubulin) were down- Engineering 4 (7) (2020) 732–742.
[6] Y. Shi, R. van der Meel, X. Chen, T. Lammers, The EPR effect and beyond: strategies
regulated. This encouraging therapeutic outcome in the PDX models to improve tumor targeting and cancer nanomedicine treatment efficacy,
could be achievable via NPs@Ce6 with two-stage irradiation against Theranostics 10 (17) (2020) 7921–7924.
bladder cancer. [7] D. Pan, X. Zheng, Q. Zhang, Z. Li, Z. Duan, W. Zheng, M. Gong, H. Zhu, H. Zhang,
Q. Gong, Z. Gu, K. Luo, Dendronized-polymer disturbing cells’ stress protection by
targeting metabolism leads to tumor vulnerability, Adv. Mater. 32 (14) (2020)
1907490.
Declaration of competing interest [8] S. Ahmadi, N. Rabiee, M. Bagherzadeh, F. Elmi, Y. Fatahi, F. Farjadian,
N. Baheiraei, B. Nasseri, M. Rabiee, N.T. Dastjerd, A. Valibeik, M. Karimi, M.
The authors declare that they have no known competing financial R. Hamblin, Stimulus-responsive sequential release systems for drug and gene
delivery, Nano Today 34 (2020) 100914.
interests or personal relationships that could have appeared to influence [9] Y. Dong, Y. Tu, K. Wang, C. Xu, Y. Yuan, J. Wang, A general strategy for
the work reported in this paper. macrotheranostic prodrug activation: synergy between the acidic tumor
microenvironment and bioorthogonal chemistry, Angew. Chem. 59 (18) (2020)
7168–7172.
Acknowledgements [10] X. Yang, C. Hu, F. Tong, R. Liu, Y. Zhou, L. Qin, L. Ouyang, H. Gao, Tumor
microenvironment-responsive dual drug dimer-loaded PEGylated bilirubin
This work was supported by the National Natural Science Foundation nanoparticles for improved drug delivery and enhanced immune-chemotherapy of
breast cancer, Adv. Funct. Mater. 29 (32) (2019) 1901896.
of China (52073193, 82002692, 51873120, 81621003), 1‧3‧5 project for [11] Y. Wu, F. Li, X. Zhang, Z. Li, Q. Zhang, W. Wang, D. Pan, X. Zheng, Z. Gu, H. Zhang,
disciplines of excellence, West China Hospital, Sichuan University Q. Gong, K. Luo, Tumor microenvironment-responsive PEGylated heparin-
(ZYJC21013), Department of Science and Technology of Sichuan Prov­ pyropheophorbide-a nanoconjugates for photodynamic therapy, Carbohydr.
Polym. 255 (2021) 117490.
ince (2018JY0574), Post-Doctor Research Project of Sichuan University [12] S. Wang, G. Yu, Z. Wang, O. Jacobson, R. Tian, L.-S. Lin, F. Zhang, J. Wang,
(2021SCU12016), and Post-Doctor Research Project, West China Hos­ X. Chen, Hierarchical tumor microenvironment-responsive nanomedicine for
pital, Sichuan University (2018JY0574, 2019HXBH057, programmed delivery of chemotherapeutics, Adv. Mater. 30 (40) (2018) 1803926.
[13] L. Luo, F. Xu, H. Peng, Y. Luo, X. Tian, G. Battaglia, H. Zhang, Q. Gong, Z. Gu,
2020HXBH094). M. Kopytynski was supported by the NIHR Imperial
K. Luo, Stimuli-responsive polymeric prodrug-based nanomedicine delivering
Biomedical Research Centre (BRC). We thank Ping Liao (Laboratory of nifuroxazide and doxorubicin against primary breast cancer and pulmonary
Anesthesia and Critical Care Medicine, Translational Neuroscience metastasis, J. Contr. Release 318 (2020) 124–135.
[14] Y. Cheng, D.L. Sellers, J.-K.Y. Tan, D.J. Peeler, P.J. Horner, S.H. Pun, Development
Center, Department of Anesthesiology of West China Hospital, Sichuan
of switchable polymers to address the dilemma of stability and cargo release in
University, Chengdu, Sichuan, China) for imaging analysis. We thank polycationic nucleic acid carriers, Biomaterials 127 (2017) 89–96.
Guang Yang, Yang Yang, Xiaoting Chen and Guangneng Liao from the [15] X. Tian, S. Angioletti-Uberti, G. Battaglia, On the design of precision
Animal Experimental Center of West China Hospital of Sichuan Uni­ nanomedicines, Science advances 6 (4) (2020), eaat0919.
[16] M.E. Vance, L.C. Marr, Exposure to airborne engineered nanoparticles in the indoor
versity for technical assistance in animal experiments. We are also environment, Atmos. Environ. 106 (2015) 503–509.
grateful to Sisi Wu, Xuemei Chen, Zhiqian Li, Guonian Zhu, Yanjing [17] S. Mishra, P. Webster, M.E. Davis, PEGylation significantly affects cellular uptake
Zhang, Hongying Chen, Yan Wang, Yu Ding, Fengfeng Chen, Li Zhou, and intracellular trafficking of non-viral gene delivery particles, Eur. J. Cell Biol.
83 (3) (2004) 97–111.
Jinkui Pi, Bo Su, Lei Wu, Jie Zhang, Shiqing Fu and Yi Zhang (Core [18] Z. Li, S. Li, Y. Guo, C. Yuan, X. Yan, K.S. Schanze, Metal-free nanoassemblies of
Facility of West China Hospital of Sichuan University) for their technical water-soluble photosensitizer and adenosine triphosphate for efficient and precise
assistance in in vitro and in vivo studies. photodynamic cancer therapy, ACS Nano 15 (3) (2021) 4979–4988.
[19] S. Mandal, G. Mann, G. Satish, A. Brik, Enhanced live-cell delivery of synthetic
proteins assisted by cell-penetrating peptides fused to DABCYL, Angew. Chem. 60
Appendix A. Supplementary data (13) (2021) 7333–7343.
[20] M. Shu, J. Tang, L. Chen, Q. Zeng, C. Li, S. Xiao, Z. Jiang, J. Liu, Tumor
microenvironment triple-responsive nanoparticles enable enhanced tumor
Supplementary data to this article can be found online at https://doi. penetration and synergetic chemo-photodynamic therapy, Biomaterials 268 (2021)
org/10.1016/j.biomaterials.2021.121061. 120574.

12
P. Tan et al. Biomaterials 277 (2021) 121061

[21] Y. Wang, G. Wei, X. Zhang, F. Xu, X. Xiong, S. Zhou, A step-by-step multiple [37] M.I. Love, W. Huber, S. Anders, Moderated estimation of fold change and
stimuli-responsive nanoplatform for enhancing combined chemo-photodynamic dispersion for RNA-seq data with DESeq2, Genome Biol. 15 (12) (2014) 550.
therapy, Adv. Mater. 29 (12) (2017) 1605357. [38] A. Butler, P. Hoffman, P. Smibert, E. Papalexi, R. Satija, Integrating single-cell
[22] L. Wu, X. Cai, H. Zhu, J. Li, D. Shi, D. Su, D. Yue, Z. Gu, PDT-driven highly efficient transcriptomic data across different conditions, technologies, and species, Nat.
intracellular delivery and controlled release of CO in combination with sufficient Biotechnol. 36 (5) (2018) 411–420.
singlet oxygen production for synergistic anticancer therapy, Adv. Funct. Mater. 28 [39] T. Stuart, A. Butler, P. Hoffman, C. Hafemeister, E. Papalexi, W.M. Mauck 3rd,
(41) (2018) 1804324. Y. Hao, M. Stoeckius, P. Smibert, R. Satija, Comprehensive integration of single-cell
[23] X. Yi, J.J. Hu, J. Dai, X. Lou, Z. Zhao, F. Xia, B.Z. Tang, Self-guiding polymeric data, Cell 177 (7) (2019) 1888–1902, e21.
prodrug micelles with two aggregation-induced emission photosensitizers for [40] G. Yu, L.-G. Wang, Y. Han, Q.-Y. He, clusterProfiler: an R package for comparing
enhanced chemo-photodynamic therapy, ACS Nano 15 (2) (2021) 3026–3037. biological themes among gene clusters, OMICS A J. Integr. Biol. 16 (5) (2012)
[24] C. Zhang, W.-J. Qin, X.-F. Bai, X.-Z. Zhang, Nanomaterials to relieve tumor hypoxia 284–287.
for enhanced photodynamic therapy, Nano Today 35 (2020) 100960. [41] A. Subramanian, P. Tamayo, V.K. Mootha, S. Mukherjee, B.L. Ebert, M.A. Gillette,
[25] Y. Cai, D. Ni, W. Cheng, C. Ji, Y. Wang, K. Müllen, Z. Su, Y. Liu, C. Chen, M. Yin, A. Paulovich, S.L. Pomeroy, T.R. Golub, E.S. Lander, J.P. Mesirov, Gene set
Enzyme-triggered disassembly of perylene monoimide-based nanoclusters for enrichment analysis: a knowledge-based approach for interpreting genome-wide
activatable and deep photodynamic therapy, Angew. Chem. 132 (33) (2020) expression profiles, Proc. Natl. Acad. Sci. U. S. A 102 (43) (2005) 15545–15550.
14118–14122. [42] V.K. Mootha, C.M. Lindgren, K.-F. Eriksson, A. Subramanian, S. Sihag, J. Lehar,
[26] A. Dalpiaz, G. Paganetto, G. Botti, B. Pavan, Cancer stem cells and nanomedicine: P. Puigserver, E. Carlsson, M. Ridderstråle, E. Laurila, N. Houstis, M.J. Daly,
new opportunities to combat multidrug resistance? Drug Discov. Today 25 (9) N. Patterson, J.P. Mesirov, T.R. Golub, P. Tamayo, B. Spiegelman, E.S. Lander, J.
(2020) 1651–1667. N. Hirschhorn, D. Altshuler, L.C. Groop, PGC-1α-responsive genes involved in
[27] S. Zhen, X. Yi, Z. Zhao, X. Lou, F. Xia, B.Z. Tang, Drug delivery micelles with oxidative phosphorylation are coordinately downregulated in human diabetes,
efficient near-infrared photosensitizer for combined image-guided photodynamic Nat. Genet. 34 (3) (2003) 267–273.
therapy and chemotherapy of drug-resistant cancer, Biomaterials 218 (2019) [43] R. Zhang, K. Luo, J. Yang, M. Sima, Y. Sun, M.M. Janát-Amsbury, J. Kopeček,
119330. Synthesis and evaluation of a backbone biodegradable multiblock HPMA
[28] X. Zhao, J. Liu, J. Fan, H. Chao, X. Peng, Recent Progress in Photosensitizers for copolymer nanocarrier for the systemic delivery of paclitaxel, J. Contr. Release 166
Overcoming the Challenges of Photodynamic Therapy: from Molecular Design to (1) (2013) 66–74.
Application, Chemical Society reviews, 2021. [44] M. Li, L. Yin, L. Wu, Y. Zhu, X. Wang, Paclitaxel inhibits proliferation and promotes
[29] Y. Yang, L. Wang, H. Cao, Q. Li, Y. Li, M. Han, H. Wang, J. Li, Photodynamic apoptosis through regulation ROS and endoplasmic reticulum stress in
therapy with liposomes encapsulating photosensitizers with aggregation-induced osteosarcoma cell, Molecular & Cellular Toxicology 16 (4) (2020) 377–384.
emission, Nano Lett. 19 (3) (2019) 1821–1826. [45] X. Ren, B. Zhao, H. Chang, M. Xiao, Y. Wu, Y. Liu, Paclitaxel suppresses
[30] M. Hidalgo, F. Amant, A.V. Biankin, E. Budinská, A.T. Byrne, C. Caldas, R. proliferation and induces apoptosis through regulation of ROS and the AKT/MAPK
B. Clarke, S. de Jong, J. Jonkers, G.M. Mælandsmo, S. Roman-Roman, J. Seoane, signaling pathway in canine mammary gland tumor cells, Mol. Med. Rep. 17 (6)
L. Trusolino, A. Villanueva, Patient-derived xenograft models: an emerging (2018) 8289–8299.
platform for translational cancer research, Canc. Discov. 4 (9) (2014) 998–1013. [46] J. Ding, G. Lu, W. Nie, L.L. Huang, Y. Zhang, W. Fan, G. Wu, H. Liu, H.Y. Xie, Self-
[31] C. Willyard, The mice with human tumours: growing pains for a popular cancer Activatable photo-extracellular vesicle for synergistic trimodal anticancer therapy,
model, Nature 560 (7717) (2018) 156–157. Adv. Mater. 33 (7) (2021), e2005562.
[32] J.J. Tentler, A.C. Tan, C.D. Weekes, A. Jimeno, S. Leong, T.M. Pitts, J.J. Arcaroli, [47] T.M. Ashton, W.G. McKenna, L.A. Kunz-Schughart, G.S. Higgins, Oxidative
W.A. Messersmith, S.G. Eckhardt, Patient-derived tumour xenografts as models for phosphorylation as an emerging target in cancer therapy, Clin. Canc. Res. 24 (11)
oncology drug development, Nat. Rev. Clin. Oncol. 9 (6) (2012) 338–350. (2018) 2482–2490.
[33] Y.S. DeRose, G. Wang, Y.C. Lin, P.S. Bernard, S.S. Buys, M.T. Ebbert, R. Factor, [48] E. Gaude, C. Frezza, Tissue-specific and convergent metabolic transformation of
C. Matsen, B.A. Milash, E. Nelson, L. Neumayer, R.L. Randall, I.J. Stijleman, B. cancer correlates with metastatic potential and patient survival, Nat. Commun. 7
E. Welm, A.L. Welm, Tumor grafts derived from women with breast cancer (2016) 13041.
authentically reflect tumor pathology, growth, metastasis and disease outcomes, [49] G. Solaini, G. Sgarbi, A. Baracca, Oxidative phosphorylation in cancer cells,
Nat. Med. 17 (11) (2011) 1514–1520. Biochim. Biophys. Acta Bioenerg. 1807 (6) (2011) 534–542.
[34] T. Murayama, N. Gotoh, Patient-derived xenograft models of breast cancer and [50] T. Finkel, Signal transduction by mitochondrial oxidants, J. Biol. Chem. 287 (7)
their application, Cells 8 (6) (2019). (2012) 4434–4440.
[35] T. Chou, N. Martin, CompuSyn for Drug Combinations: PC Software and User’s [51] A.L. Smith, T.P. Robin, H.L. Ford, Molecular pathways: targeting the TGF-beta
Guide: a Computer Program for Quantitation of Synergism and Antagonism in Drug pathway for cancer therapy, Clin. Canc. Res. 18 (17) (2012) 4514–4521.
Combinations, and the Determination of IC50 and ED50 and LD50 Values, [52] Y. Wu, B.P. Zhou, TNF-alpha/NF-kappaB/Snail pathway in cancer cell migration
ComboSyn, Paramus, NJ, 2005. and invasion, Br. J. Canc. 102 (4) (2010) 639–644.
[36] A. Dobin, C.A. Davis, F. Schlesinger, J. Drenkow, C. Zaleski, S. Jha, P. Batut, [53] B. Muz, P. de la Puente, F. Azab, A.K. Azab, The role of hypoxia in cancer
M. Chaisson, T.R. Gingeras, STAR: ultrafast universal RNA-seq aligner, progression, angiogenesis, metastasis, and resistance to therapy, Hypoxia (2015)
Bioinformatics 29 (1) (2013) 15–21. 83.

13