Abstract

   Tumor protein 53 (TP53) mutation in bladder carcinoma (BC), upregulates
   the transcription of carbamoyl phosphate synthetase 1 (CPS1), to reduce
   intracellular ammonia toxicity. To leverage ammonia combating BC, here,
   an intravesically perfusable nanoporter‐encased hydrogel system is
   reported. A biomimetic fusogenic liposomalized nanoporter (FLNP) that
   is decorated with urea transporter‐B (UT‐B) is first synthesized with
   protonated chitosan oligosaccharide for bladder tumor‐targeted
   co‐delivery of urease and small interfering RNA targeting CPS1
   (siCPS1). Mussel‐inspired hydrogel featured with dual functions of
   bio‐adhesion and injectability is then fabricated as the reservoir for
   intravesical immobilization of FLNP. It is found that FLNP‐mediated
   UT‐B immobilization dramatically induces urea transportation into tumor
   cells, and co‐delivery of urease and siCPS1 significantly boosts
   ammonia accumulation in tumor inducing cell apoptosis. Treatment with
   hybrid system exhibits superior anti‐tumor effect in orthotopic bladder
   tumor mouse model and patient‐derived xenograft model, respectively.
   Combined with high‐protein diet, the production of urinary urea
   increases, leading to an augmented intracellular deposition of ammonia
   in BC cells, and ultimately an enhanced tumor inhibition. Together, the
   work establishes that cascade modulation of ammonia in tumor cells
   could induce tumor apoptosis and may be a practical strategy for
   eradication of TP53‐mutated bladder cancer.

   Keywords: bladder carcinoma, carbamoyl phosphate synthetase 1,
   metabolic modulation, urea, urea transporter‐B
     __________________________________________________________________

   Fusogenic liposomalized nanoporter (FLNP)‐mediated urea transporter‐B
   immobilization dramatically enhances urea transport into tumor cells,
   and codelivery of urease and small interfering RNA targeting carbamoyl
   phosphate synthetase 1 significantly boosts ammonia accumulation in
   tumor inducing cell apoptosis. A mussel‐inspired hydrogel is fabricated
   as the reservoir for intravesical immobilization of FLNP.

   graphic file with name ADVS-10-2206893-g006.jpg

1. Introduction

   Bladder carcinoma (BC) is the most common neoplasm of the urinary
   tract, with ≈573 000 new cases and 213 000 deaths per year, globally.^[
   [44]^1 ^] Non‐muscle‐invasive bladder cancer (NMIBC) accounts for ≈70%
   of all bladder cancers.^[ [45]^2 ^] Despite intensive treatment with
   intravesical instillation of chemotherapeutics and radical cystectomy
   following transurethral tumor resection, two‐thirds of patients with
   NMIBC will suffer recurrence and even disease progression to
   muscle‐invasive bladder cancer, among which over ≈55% of patients die
   within 5 years regardless any of additional treatments.^[ [46]^3 ^] It
   is imperative to develop effective strategies for targeting and
   eliminating the refractory neoplastic burden within the bladder, which
   remains elusive.

   Mutation of tumor protein 53 (TP53) is the most frequent genetic
   alteration in bladder cancer, which results in transcriptional
   activation, senescence, apoptosis, and changes in metabolism,^[ [47]^4
   , [48]^5 ^] promoting tumor progression (Figure [49]1a). Carbamoyl
   phosphate synthetase 1 (CPS1), a multidomain mitochondrial
   rate‐limiting enzyme protein, catalyze the first committed step of the
   urea recycle by converting ammonia into carbamoyl phosphate for ammonia
   detoxification and disposal.^[ [50]^6 ^] We found that in TP53‐mutant
   BC, the CPS1 expression level was significantly upregulated
   (Figure [51]1b). Moreover, stratification of patients by CPS1
   expression revealed a disease free and overall survival advantage for
   patients with lower CPS1 expression (Figure [52]1c,d). Consistent with
   the bioinformatic data, CPS1 protein levels in TP53‐mutant tumor
   tissues were also considerably elevated compared with those in
   TP53‐wild‐type tumor tissues in BC patients (Figure [53]1e). These data
   indicate that TP53‐mutant bladder cancer considerably upregulates the
   transcription of CPS1 to promote ammonia metabolism in cancerous cells,
   which may otherwise be toxic to them. These results have thus inspired
   us to conclude that harnessing the intracellular accumulation of
   ammonia might be an amenable approach to eradicate BC.

Figure 1.

   Figure 1
   [54]Open in a new tab

   CPS1 expression is almost always increased together with TP53 mutation
   in bladder cancer. a) Primary genetic alterations in bladder cancer
   patients. b) Quantification of CPS1 mRNA levels in human TP53 wild‐type
   (n = 50) and TP53‐mutant (n = 50) tumor tissue samples. Student's
   t‐test. Curves showing the c) disease‐free survival and d) overall
   survival of CPS1 high (n = 80) and low (n = 81) expression bladder
   cancer patients. The number of n indicates biologically independent
   samples. e) Two representative HE and IHC images of CPS1 expression in
   human TP53 wild‐type and TP53‐mutant tumor tissue samples. Scale bars
   in low magnification images represent 500 µm, and in high magnification
   images represent 100 µm. f) Schematic illustration of the intravesical
   instillation of the nanoparticle‐hydrogel system in bladder cancer that
   initiates the accumulation of ammonium ions in the cells and damage to
   mitochondria and ultimately leads to the apoptosis of tumor cells. Data
   are presented as mean ± s.d. ***p < 0.001 by Student's t‐test.

   Bladder tumors usually originate from the urothelium and are directly
   exposed to concentrated urinary urea,^[ [55]^7 ^] which can be
   catalyzed into cytotoxic ammonia via urease. Orientated polarization of
   urea toward tumors is first and foremost essential for modulating
   cytotoxic ammonia for combating BC. Urea transporter‐B (UT‐B), a
   passive membrane channel, is highly expressed in the membrane of
   erythrocytes and some epithelial and endothelial cells, for example,
   14 000 UT‐B per erythrocyte.^[ [56]^8 , [57]^9 ^] UT‐B facilitates
   highly efficient transportation of urea.^[ [58]^10 ^] Artificially
   re‐engineered membranes with UT‐B may be a feasible approach to promote
   urea transportation in bladder tumor cells. Therefore, small
   interfering RNA targeting CPS1 (siCPS1) co‐delivered with urease may
   synergistically increase the intracellular deposit of ammonia to a
   level of cellular toxicity against cancerous cells.

   Intravesical instillation is the standard‐of‐care treatment for bladder
   cancer.^[ [59]^11 ^] Bladder tumors can be directly connected to the
   outside of the body through the urethra, which provides favorable
   conditions for the development of innovative drugs for the treatment of
   bladder cancer through intravesical instillation.^[ [60]^12 ^] To
   enhance intravesical drug retention, we sought to construct an
   instilling nanoparticle‐hydrogel supersystem for BC treatment. To
   cascade boost the deposit of intracellular ammonia, a biomimetic
   fusogenic liposomalized nanoporter (FLNP) that was decorated with UT‐B
   was first synthesized with protonated chitosan oligosaccharide (COS)
   for bladder tumor‐targeted co‐delivery of urease and siCPS1. A
   mussel‐inspired hydrogel featured with dual functions of bio‐adhesion
   and injectability was then fabricated as a reservoir for intravesical
   immobilization of FLNPs. After intravesical instillation, the adhesive
   hydrogel adhered to the wet bladder wall. Bld‐1 is a specific targeting
   peptide for bladder cancer cells, and its affinity for bladder cancer
   cells is far higher than that for other cells.^[ [61]^13 , [62]^14 ^]
   The FLNPs liberated from the gel tracked and actively targeted the
   bladder tumor cells via the Bld‐1 peptide. Subsequently, in a membrane
   fusion manner, UT‐B was immobilized in cancerous cell membranes. The
   unshelled nanoporter (NP) then disintegrated, and the cargo, including
   urease and siCPS1, were released into the cytosol. We found that
   FLNP‐mediated UT‐B‐cell membrane immobilization dramatically enhanced
   urea transport into tumor cells, and co‐delivery of urease and siCPS1
   significantly boosted the intracellular accumulation of ammonia to an
   eradicating level against bladder cancer cells (Figure [63]1f). The
   multipotencies of the instilling nanoparticle‐hydrogel system were
   explored by modulation of the metabolic cascade of intracellular
   ammonia to combat BC in multiple orthotopic bladder tumor models.

2. Results and Discussion

   COS is an inexpensive and ideal cationic biopolymer material for drug
   delivery due to its biocompatibility, noncytotoxicity and
   biodegradability.^[ [64]^15 ^] It has been classified by the FDA as
   generally recognized as safe for humans.^[ [65]^16 ^] Inspired by this,
   we used the preprotonated COS to assemble with urease and form an NP
   for efficient protein delivery (Figure [66]2a). To construct the NP, we
   first prepared COS‐loaded (COSlated) urease by mixing COS (0.2 g mL^−1)
   solutions with urease, resulting in NPs with an average size of
   64.34 nm, and a uniform shape (Figure [67]S1a,b, Supporting
   Information). The prepared COSlated urease is positively charged, with
   a zeta potential of 22.21 mV (Figure [68]2b). The COSlated urease was
   then used to condense the negatively charged siCPS1 at different weight
   ratios and examined on a 2% agarose gel. As shown in Figure [69]S2,
   Supporting Information, the COSlated urease completely condensed the
   siRNA at a weight ratio of 8:1, and this result was confirmed by the
   zeta potential result in Figure [70]2b, with an average diameter of
   90.78 nm (Figure [71]2c–e). Additionally, we further investigated the
   changing of urease activity before and after being stirred with COS,
   and found that there was no significant changing before and after
   urease being stirred with COS (Figure [72]S3, Supporting Information).
   We then coated the NPs with a biomimetic fusogenic liposome (BFL) for
   urea transport by reconstituting UT‐B into the liposome, while the
   Bld‐1 peptide‐decorated lipid membrane greatly improved the targeting
   capacity of the NP. Transmission electron micrographs revealed that
   FLNPs had a spherical vesicular structure with a diameter of ≈111.08 nm
   (Figure [73]2f,g), followed by conflation with the fusogenic liposome.
   Through urease ELISA assay and siRNA assay, we further confirmed the
   co‐loading of urease and CPS1 siRNA in the FLNP. As shown in Figure
   [74]S4a,b, Supporting Information, the ELISA and gel assay showed that
   there were urease and CPS1 siRNA loading in FLNP. According to the drug
   loading formula, we calculated the drug loading of urease and siRNA in
   FLNP is 6.94% ± 0.39% and 1.29% ± 0.06% respectively. The presence of
   UT‐B on the BFL membrane was verified by western blotting and
   fluorescent antibody detection (Figure [75]2h and Figure [76]S4c,
   Supporting Information).

Figure 2.

   Figure 2
   [77]Open in a new tab

   Synthesis procedure and characterization of the intravesical
   instillation superstructure. a) Schematic illustration showing FLNPs
   preparation. b) Zeta potential of each formulation (n = 4). c) TEM
   images of NPs. Scale bar indicates 100 nm. d) Dynamic light scattering
   (DLS) analysis of NPs. e) Particle size distributions of each
   formulation (n = 4). f) TEM images of FLNPs. Scale bar, 100 nm. g) DLS
   analysis of FLNPs. h) Western blot analysis of UT‐B in different
   formulations. i) Schematic illustration of the bionic principle of
   HA‐DA. j) Images showing the gelator solution before gelation (left)
   and the hydrogel after gelation (right). k) SEM (scale bar, 20 µm)
   images of the hydrogel. l) Rheology traces of the HA‐DA hydrogels. m)
   Schematic diagram of ultrasound‐guided intravesical perfusion of the
   hydrogel. n) The image of bladder dissection showing that the hydrogel
   adheres to the bladder wall at 1 h after bladder irrigation, scale bar:
   1 cm.

   Mytilus is a marine organism that can adhere to wet rock walls. The
   functionalization of biopolymers containing catechol in bionic mussels
   has been used to improve the biocompatibility and enhance the adhesion
   of 3D hydrogels.^[ [78]^17 ^] The proteins that play an adhesive role
   in mussels include repeated catecholic amino acids, lysine residues,
   and 3,4‐dihydroxy‐L‐phenylalanine (L‐DOPA, DA).^[ [79]^18 ^] Hyaluronic
   acid‐DA (HA‐DA) hydrogel can efficiently adhere to the surface of
   tissues. This ability mainly depends on the high affinity between the
   DA group in HA‐DA hydrogel and the diversity nuclei (e.g., amines,
   thiol, and imidazole), so that the DA group can bind to the peptides
   and proteins on the surface of tissues (Figure [80]2i). As shown in
   Figure [81]S5, Supporting Information, there is a catechol
   proton‐specific peak at ≈7 ppm in the nuclear magnetic resonance (NMR)
   assay result, which demonstrates the conjugation of dopamine to HA. As
   shown in Figure [82]2j, the color of HA‐DA pregel solution changes from
   transparent color to brown immediately during gelation due to the
   oxidative conversion of the catechol moiety into their corresponding
   o‐quinones. We further investigated the modulus changing during the
   process of hydrogel formation from the liquid phase to the solid phase.
   As shown in Figure [83]S6, Supporting Information, the cross‐linkers
   were added after a 3.5 min conditioning step, and the gel points (G′ =
   G″) were rapidly reached. As shown in Figure [84]2k, the HA‐DA
   hydrogels exhibited a homogeneous porous structure. Afterward, dynamic
   mechanical analysis was used for determined the HA‐DA hydrogel (2%)
   rheological properties after gelation. We quantified the modulus of
   hydrogel in a frequency range from 0.1 to 1 Hz, and found that the
   storage modulus (G′) was constantly higher than the loss modulus (G″),
   demonstrating the stability and viscoelasticity of the HA‐DA hydrogel
   (Figure [85]2l). To form the FLNP@HA‐DA hydrogel, we loaded 5 mg mL^−1
   FLNP in the hydrogel. We also tested the FLNP‐hydrogel structure and
   release behavior. The SEM result clearly show the liposomes loaded on
   hydrogel (Figure [86]S7a, Supporting Information), and the release
   experiment indicated that the FLNP sustained release from the hydrogel
   up to 24 h (Figure [87]S7b, Supporting Information). We further
   explored the injectability of the hydrogel. As shown in Figure [88]S8a,
   Supporting Information, the hydrogel passed through an 18G syringe
   needle smoothly and forming various designed shapes. Furthermore, we
   detected the content of FLNP in the hydrogel before and after injection
   by using the method of Dil‐labeled FLNP, and found that the injection
   had no significant effect on the content of FLNP in the hydrogel
   (Figure [89]S8b, Supporting Information). To verify the adhesion of the
   hydrogel to the bladder wall of large animals, we used a pig model.
   Under the guidance of ultrasound, hydrogel was injected into the
   bladder of a pig. After 1 h, the bladder was dissected out of the body
   and cut open; photograph showed the status of the hydrogel in the
   bladder wall (Figure [90]2m,n). Additionally, we further investigated
   the adhesion time of the hydrogel to the bladder wall. As shown in
   Figure [91]S9, Supporting Information, at 24 h after perfusion, the
   hydrogel adhered to the bladder wall decreased compared with that at 1
   h. And at 48 h, the hydrogel adhered to the bladder wall was
   significantly reduced to almost completely degraded. This result
   indicated that the hydrogel can continuously adhere to the bladder wall
   for at least 24–48 h.

   In order to verify the therapeutic effect of our designed FLNP on
   bladder cancer, we first evaluated the targeting capability of the
   targeted biomimetic FLNP to bladder cancer cells. To confirm this
   feature, the cellular uptake of Bld‐1 peptide‐modified FLNPs was
   explored in bladder cancer (MB49) cells. As a control, FLNPs without
   the targeting ligand were added to MB49 cells and cultured under
   similar conditions. After 4 h of incubation with Dil‐labeled cancer
   cells, the Bld‐1 peptide‐modified FLNPs exhibited a much higher UT‐B
   fluorescence compared to the slight fluorescence intensity of the FLNPs
   without Bld‐1 peptide group. Moreover, the green signal from UT‐B in
   the FLNP group gradually merged with the red circle, indicating that
   FLNPs could fuse with the MB49 cell membrane (Figure [92]3a). For the
   FLNP without Bld‐1 group, although the FLNPs quickly adhered to the
   cell membrane, fusion was too rare to be observed (Figure [93]3a),
   which further confirmed the capability of the active targeting of the
   Bld‐1 peptide in bladder cancer cells. Furthermore, the different
   cellular uptake induced by diverse formulations was quantitatively
   investigated by flow cytometry analysis. The cellular uptake of Bld‐1
   peptide‐modified FLNPs was increased compared with that of FLNPs
   without Bld‐1 (Figure [94]S10, Supporting Information), indicating that
   the Bld‐1 peptide significantly improved the cellular enrichment of
   FLNPs due to the preserved tumor‐homing ability of the Bld‐1 peptide.
   We subsequently confirmed the membrane fusion pathway of the FLNP
   membrane by confocal laser scanning microscopy (CLSM). When we used the
   lysosome‐specific probe LysoTracker Red and UT‐B fluorescent antibody
   to label the cells, the results of confocal microscopy showed that UT‐B
   and lysosomes rarely co‐localized in the cells, indicating that bladder
   cancer cells did not through endocytosis to uptake FLNP. However, when
   we replace the DMPC in FLNP with DSPC, we observed endocytosis and
   colocalization between the lysosomes and the UT‐B, which further
   confirmed the fusogenic potential of the FLNP membrane with bladder
   cancer cells (Figure [95]3b). As shown in Figure [96]S11, Supporting
   Information, urease and CPS1 siRNA were initially colocalized after the
   nanoparticles entered the cytoplasm. After 6 h, the colocalization of
   urease and CPS1 siRNA decreased, and they were distributed in the
   cytoplasm, indicating that urease and CPS1 siRNA were released into the
   cytoplasm. To investigate the capability of FLNPs to deliver siCPS1
   into tumor cells and silence CPS1 gene expression, we next examined the
   cellular gene silencing efficiency of FLNPs in vitro. After 24 h of
   incubation with different formulations, we used western blotting to
   analyze CPS1 and UT‐B expression in bladder cancer cells. As shown in
   Figure [97]3c, FLNP‐delivered siCPS1 dramatically downregulated the
   levels of CPS1 in MB49 cells at a siCPS1 concentration of 1.0 µg mL^−1,
   while CPS1 expression in cancer cells treated with the NPs was still
   evident. These results further indicate that FLNPs can transfer siRNA
   into tumor cells and silence the CPS1 gene with satisfactory efficacy.
   To better enhance the killing effect of ammonia metabolism blockade on
   tumor cells, we not only interfered with key enzymes of ammonia
   metabolism, but also increased the content of urea channels on the
   tumor cell membrane by means of membrane fusion liposome delivery.
   After the nanoparticles were administered, UT‐B on the membrane fusion
   liposome fused with the cell membrane of bladder cancer. After fusion,
   UT‐B was highly expressed on bladder cancer cells (Figure [98]3c),
   which provided a channel for the rapid passage of high concentrations
   of urinary urea to the inside of tumor cells.

Figure 3.

   Figure 3
   [99]Open in a new tab

   FLNP mediated high levels of UT‐B on the cell membrane and CPS1
   inhibition of bladder cancer cells. a) Cellular uptake of FLNP with
   Bld‐1 or FLNP without Bld‐1 by bladder cancer cells as measured through
   confocal microscopy. Images from left to right show cell membrane
   (red), UT‐B (green), DAPI‐stained cell nuclei (blue) and merged images.
   Scale bar indicates 20 µm. b) Images showing the cell components of
   bladder cancer cells with FLNP‐DMPC and FLNP‐DSPC treatment. The red
   signal indicates lysosome, green signal indicates UT‐B, and blue signal
   indicates nuclei. Scale bar represents 20 µm. c) Western blot showing
   CPS1 and UT‐B protein expression in bladder cancer cells after
   different treatments for 24 h. G1: control; G2: COSlated urease; G3:
   NP; G4: NP‐laden fusogenic liposome; G5: NP‐laden UT‐B reconstituted
   fusogenic liposome; G6: FLNP. d,e) Apoptosis analysis of bladder cancer
   cells after 24 h of different treatments in culture medium supplemented
   with FBS and urea (100 mm) using cytometry (n = 3 independent
   experiments). f) Images of bladder cancer cells after 24 h of different
   treatments in culture medium supplemented with FBS and urea (100 mm).
   Scale bar indicates 100 µm. g) Fluorescence images of bladder cancer
   cells in live/dead staining experiments after 24 h of different
   treatments in culture medium supplemented with FBS and urea (100 mm).
   Scale bars indicate 150 µm. d–g) G1: control; G2: COS; G3: COSlated
   urease; G4: NP; G5: NP‐laden fusogenic liposome; G6: NP‐laden UT‐B
   reconstituted fusogenic liposome; G7: FLNP. Data are presented as mean
   ± s.d. **p < 0.01; ****p < 0.0001 by one‐way ANOVA test.

   As shown in Figure [100]3d,e, the cells treated with NPs showed ≈42.40%
   cell apoptosis, and the apoptosis rate of cells treated with NP‐laden
   UT‐B‐reconstituted fusogenic liposomes increased to 76.85%, while FLNPs
   induced up to 93.51% cell apoptosis. For the cell viability assay, only
   moderate anticancer activity was observed in the COS or COSlated urease
   group, while the NPs exhibited a higher death rate than the COSlated
   urease, indicating that the siRNA can increase ammonia accumulation,
   leading to a synergistic therapeutic effect (Figure [101]3f). In
   addition, the cells treated with the NP‐laden UT‐B reconstituted
   fusogenic liposome showed a higher cell proliferation inhibition effect
   compared with NP‐laden fusogenic liposome (Figure [102]S12a,b,
   Supporting Information), suggesting that UT‐B reconstituted fusogenic
   liposome on the NP may increase the urease cellular uptake and
   subsequently enhance the ammonia concentration. This suggested that the
   antitumor efficacy was greatly enhanced with the increased
   intracellular ammonia concentration due to the tumor‐selective
   targeting of Bld‐1. Moreover, a live/dead viability/cytotoxicity assay
   for fluorescence microscopy analysis was performed to reveal the
   cytotoxicity mechanism. As shown in Figure [103]3g, FLNPs induced
   significant death in bladder cancer cells with urea supplements
   (100 mm). Additionally, we investigated the toxicity of FLNP to normal
   cells. Compared with bladder cancer cells, FLNPs were significantly
   less toxic to the normal urothelial cell line SV‐HUC‐1 (Figure
   [104]S13, Supporting Information).

   To investigate the underlying mechanism of the FLNPs effects on bladder
   cancer progression, we first determined ammonia accumulation at the
   whole protein level. An examination was performed on proteins
   differentially expressed in tumor cells treated with FLNPs compared to
   those treated with PBS. A total of 38 upregulated proteins and 23
   downregulated proteins were filtered out by setting the t test p‐value
   < 0.05 and the cut‐off value of log2 (fold change) >1 (Figure [105]4a).
   Enrichment of the differentially expressed proteins was investigated by
   pathway database analysis and ranked by p‐value. GO enrichment analysis
   indicated that most of these metabolites were associated with apoptosis
   process (Figure [106]4b). Through pathway enrichment analysis, we found
   that after FLNP treatment, certain signaling pathways including
   mitochondrial membrane permeability, apoptotic chromosome condensation
   and apoptotic signaling pathways, were significantly changed. As shown
   in Figure [107]4c, the results of organelle localization analysis of
   the altered proteins showed that the altered proteins were mainly
   concentrated in the mitochondria and cytoplasm after treatment with
   FLNP. Mitochondria are not only the main organelles supplying energy to
   the cell, but also the key organelles involved in ammonia metabolism in
   the cell, playing an important role in the regulation of ammonia
   metabolism in the cell.^[ [108]^19 , [109]^20 ^] To further verify the
   proteomic effects of FLNPs on mitochondria, we further examined the
   intracellular changes in mitochondria by CLSM. JC‐1 staining is a
   classical method for displaying mitochondrial membrane potential.^[
   [110]^21 ^] As shown in Figure [111]S14, Supporting Information,
   mitochondrial membrane potential was significantly decreased (cell
   staining changes from red to green) in bladder cancer cells after being
   treated with FLNPs, indicating remarkable mitochondrial damage by
   ammonia. Tom‐20 is located at the outer membrane of mitochondria to
   facilitate the introduction of proteins across the outer membrane of
   mitochondria, and is a key protein in the executive function of
   mitochondria.^[ [112]^22 ^] Our results showed that FLNPs could
   significantly reduce Tom‐20 expression in bladder cancer cells,
   indicating that nanoparticles can strongly damage the mitochondria of
   bladder cancer cells (Figure [113]4d and Figure [114]S15, Supporting
   Information). Moreover, FLNPs also decreased the expression of MCL‐1
   and BCL‐2 in tumor cells (Figure [115]4e), which are involved in
   mitochondria‐mediated apoptosis. The ERK1/2 signaling pathway controls
   diverse primary cellular processes such as cell survival,
   proliferation, fate determination, and stress responses.^[ [116]^23 ^]
   Aberrant ERK1/2 signaling underlies a wide range of important diseases
   in humans, including aging and cancer.^[ [117]^24 ^] In the present
   study, we revealed that p‐ERK1/2 protein expression was significantly
   decreased by treatment with FLNPs (Figure [118]4e). Taken together, our
   results revealed that FLNPs could induce the apoptosis of bladder
   cancer cells by disturbing the mitochondria/ERK1/2 signaling pathway.

Figure 4.

   Figure 4
   [119]Open in a new tab

   FLNPs induced tumor cell apoptosis mainly by mitochondrial destruction.
   a) Hierarchical clustering analysis based on protein expression of
   bladder cancer cells treated with FLNPs compared with the PBS treatment
   group after 24 h (n = 3). b) Significantly enriched GO term histogram
   (FLNP treated vs PBS) (top 10). c) Subcellular localization analysis of
   differentially expressed proteins. d) Confocal images of staining for
   the mitochondrial structural protein Tom‐20 after different treatments
   for 24 h (scale bar, 10 µm). e) MCL‐1, BCL‐2, and p‐ERK1/2 protein
   expression levels in bladder cancer cells after different treatments
   for 24 h. f) FLNPs inhibit cellular oxygen consumption in bladder
   cancer cells with urea supplements (100 mm). After a 24 h incubation
   with FLNPs or control agents, the cellular OCR was measured by a
   Seahorse analyzer (mean ± s.d., n = 5). g) Relative cellular ATP level
   after 24 h of different treatments in culture medium supplemented with
   FBS and urea (100 mm) (mean ± s.d., n = 5). h) Confocal analysis of
   cellular F‐actin stained by phalloidin after 24 h of different
   treatments in culture medium supplemented with FBS and urea (100 mm).
   The scale bar indicates 10 µm. i) Schematic illustration of the FLNPs
   destroying mitochondria that lead to bladder cancer cell apoptosis.
   d–h) G1: control; G2: COSlated urease; G3: NP; G4: NP‐laden fusogenic
   liposome; G5: NP‐laden UT‐B reconstituted fusogenic liposome; G6: FLNP.
   Data are presented as mean ± s.d. *p < 0.05; **p < 0.01; ***p < 0.001;
   ****p < 0.0001 for one‐way ANOVA test.

   We further examined the oxygen consumption rate (OCR) and ATP levels in
   bladder cancer cells after different treatments. As shown in
   Figure [120]4f,g, the relative percentages of OCR and ATP production of
   bladder cancer cells in the FLNP‐treated group were the lowest compared
   to other groups, which is consistent with previous findings. Moreover,
   ATP production in MB49 cells also fell sharply in the FLNP group
   (Figure [121]4g), demonstrating an efficient mitochondrial respiratory
   depression effect. The cytoskeleton carries out multiple functions
   including assisting cell movement and cell morphology changing,
   connecting the cell biochemically and physically to the external
   environment and spatially organize the contents of the cell.^[ [122]^25
   ^] We used phalloidin to stain the cytoskeleton in bladder cancer cells
   and found that the cytoskeleton was significantly damaged and
   disordered after FLNP treatment (Figure [123]4h). Mitochondria produce
   up to ninety percent of ROS in living cells,^[ [124]^26 ^] and our data
   showed that the bladder cancer cells in the FLNP‐treated group
   generated more ROS than those in the other groups (Figure [125]S16,
   Supporting Information). Together, these results indicated that FLNP
   could efficiently induce mitochondrial damage and inhibit the OXPHOS
   metabolic pathway, resulting in decreased cellular oxygen consumption
   and ATP production of bladder cancer cell (Figure [126]4i).

   Based on the encouraging results that FLNP induced bladder cancer cell
   apoptosis by ammonia accumulation, we further explored the therapeutic
   effect of hydrogel‐loaded FLNP on bladder cancer after irrigating into
   the bladder using the orthotopic MB49 bladder cancer mouse model
   (Figure [127]5a). As shown in Figure [128]5b, we used the IVIS lumina
   system to monitor tumor growth in different treatment groups of bladder
   cancer‐bearing mice. Quantitative analysis of bioluminescence signals
   showed that the tumor‐bearing mice in the FLNP hydrogel treatment group
   had the lowest tumor fluorescence intensity compared with the other
   five groups, indicating the strongest inhibitory effect on bladder
   cancer (Figure [129]5c and Figure [130]S17, Supporting Information).
   Although the average BLIS was slightly lower than that of the PBS
   control group, no significant difference was obtained in the COSlated
   urease group (Figure [131]5c). Moreover, the average BLIS for the
   NP‐laden UT‐B reconstituted fusogenic liposome group and the FLNP group
   were 25.7 and 1.2 times larger than the original tumors, respectively.
   After FLNP‐hydrogel treatment, the tumors also showed the lowest weight
   and size, consistent with the ultrasound imaging results in vivo
   (Figure [132]5d–f). In addition, the results of Ki67 staining of tumor
   sections showed that the Ki67‐positive cells of mouse tumors in the
   FLNP hydrogel treatment group were lowest, indicating that the
   proliferation of bladder cancer was dramatically inhibited
   (Figure [133]5f). In terminal deoxynucleotidyl‐transferase‐mediated
   dUTP nick‐end labeling (TUNEL)‐stained bladder tumor sections, the most
   severe apoptosis was detected in the FLNP hydrogel treatment group
   (Figure [134]5g). Therefore, benefiting from the remarkable effect of
   shrinking tumor, the survival time of bladder cancer bearing mice
   treated with FLNP hydrogel was prominently prolonged (Figure [135]5h).
   Together, these results demonstrated that FLNP‐mediated ammonium
   accumulation exhibited superior antitumor activity. In addition, there
   were no significant pathological changes in the main organs of the mice
   with different drugs treatment compared with the mice from control
   group (Figure [136]S18a, Supporting Information). It is worth
   mentioning that the peripheral blood biochemical indicators also showed
   that no significant differences were found in the levels of glutamic
   pyruvate transaminase (ALT), aspartate transaminase (AST), glucose
   (GLU), UREA, creatinine (CREA) and alkaline phosphatase (ALP) in the
   serum of mice after diverse treatments (Figure [137]S18b, Supporting
   Information). Both of these results preliminarily revealed the good
   biocompatibility of the FLNPs and validated the biosafety of the
   nanoparticle‐hydrogel treatment system.

Figure 5.

   Figure 5
   [138]Open in a new tab

   Antitumor efficacy of the intravesical instillation
   nanoparticle‐hydrogel system in an orthotopic bladder cancer mouse
   model. a) Diagram showing the experimental procedure in vivo. b)
   Bioluminescent images and c) quantified diagram of tumor‐bearing mice
   with different treatment (n = 5). Statistical significance was
   calculated using a Two‐way ANOVA. d) Images of bladder tumors harvested
   from mice with different treatment (n = 5 in G1 and G2; n = 6 in
   G3–G6). Scale bar indicates 1 cm. e) Statistical diagram of tumor
   weight (mean ± s.d., n = 5 in G1 and G2; n = 6 in G3–G6). f) Tumor
   images from left to right showing the ultrasonography, H&E‐staining
   slides and immunofluorescence which red signal indicates Ki67, green
   signal indicates WGA and blue signal indicates DAPI. Scale bars in
   ultrasonography represent 2 mm, H&E staining represent 500 µm, and
   immunofluorescence images represent 10 µm. g) Immunofluorescence images
   showing TUNEL staining of bladder tumors. Scale bar represents 20 µm.
   h) Kaplan–Meier survival analysis of the bladder cancer mouse model
   treated with different formulations (n = 10). log‐rank (Mantel–Cox)
   test. b–h) G1: control; G2: COSlated urease with hydrogel; G3: NP with
   hydrogel; G4: NP‐laden fusogenic liposome with hydrogel; G5: NP‐laden
   UT‐B reconstituted fusogenic liposome with hydrogel; G6: FLNP with
   hydrogel. Data are presented as mean ± s.d. *p < 0.05; **p < 0.01;
   ****p < 0.0001 for two‐way ANOVA test, one‐way ANOVA test and
   Kaplan–Meier survival analysis.

   Next, we further investigated whether intravesical irrigation of FLNP
   could exert an effective antitumor effect in female Sprague–Dawley rat
   models of the chemical carcinogen N‐methyl‐N‐nitrosourea (MNU) induced
   orthotopic bladder cancer (Figure [139]6a). We used ultrasound imaging
   of the rat bladder to monitor the development of tumors in the bladder
   of rats in different treatment groups. As shown in figure [140]6b, it
   summarized the effects of different treatment drugs on the orthotopic
   rat bladder tumors. During a period of 12 weeks, tumors in all groups
   developed gradually, and up to 24 weeks, it was clearly shown in the
   ultrasound imaging results that nearly half of the bladder inner space
   was occupied by diffuse bladder tumors. Comparatively, the tumors
   significantly shrunk in the other two groups, including the NP‐laden
   UT‐B reconstituted fusogenic liposome and FLNP groups. COSlated urease
   and NPs exhibited relatively weak inhibitory effect on bladder cancer.
   Excitingly, ultrasound imaging showed that the tumors in the FLNP
   treatment group were almost completely eliminated. In order to more
   intuitively verify the effect of FLNP on bladder tumors, we dissected
   the bladder from the sacrificed rats and cut the bladder open to better
   observe the tumors on the inner wall. As shown in Figure [141]6b, large
   numbers of tumors can be clearly seen on the inner wall of bladder from
   the control group rats, while there could hardly observed tumors on the
   inner wall of the bladder from the FLNP administration group, which
   further confirmed the therapeutic effect of FLNP on bladder cancer.
   Importantly, we also found severe renal metastasis in the MNU induced
   orthotopic rat bladder cancer model treated with PBS, whereas no
   metastatic tumors in kidney after treating with FLNPs (Figure [142]6b).
   As shown in Figure [143]6c, the survival time of FLNP treated rats with
   bladder cancer was significantly longer than that of other treatment
   groups. The above results indicated that the FLNP treatment showed
   superiority anti‐tumor and anti‐metastasis performance.

Figure 6.

   Figure 6
   [144]Open in a new tab

   A high protein diet enhances the antitumor effect of the intravesical
   instillation nanoparticle‐hydrogel system in bladder cancer mouse
   models. a) Schematic illustration of the experimental design in the rat
   bladder cancer model induced by MNU. b) Ultrasound images and
   macroscopic images of excised bladder‐ and kidney‐bearing tumors. c)
   Kaplan–Meier survival analysis of tumor‐bearing rats in different
   groups (n = 10). log‐rank (Mantel–Cox) test. d) Schematic illustration
   of the experimental design in the MB49 tumor‐bearing mouse model
   treated with the intravesical instillation superstructure and different
   dietary protein levels. e) Urea concentration in the urine of mice fed
   a normal or high protein diet (n = 6). Student's t‐test. f)
   Bioluminescent images and g) quantified diagram of tumor‐bearing mice
   with different treatment (n = 5), including control group, FLNP
   hydrogel with normal diet group and FLNP hydrogel with high protein
   diet group. Statistical significance was calculated via two‐way ANOVA.
   h) Images of bladder tumors harvested from mice with different
   treatment (n = 6). Scale bar indicates 1 cm. i) Statistical diagram of
   tumor weight from mice with different treatment (mean ± s.d., n = 6).
   Student's t‐test. j) Kaplan–Meier survival analysis of bladder cancer
   mice model treated with intravesically instilled FLNP‐hydrogel in
   normal or high protein diet (n = 6). log‐rank (Mantel–Cox) test. k)
   Diagram showing the experimental procedure of PDX mouse model. l)
   Ultrasound images of orthotopic bladder tumors in a PDX mouse model
   treated with intravesically instilled FLNP‐hydrogel and a normal or
   high protein diet. Scale bars, 2 mm. m) Kaplan–Meier survival analysis
   of orthotopic bladder tumors in a PDX mouse model treated with the
   intravesically instilled nanoparticle‐hydrogel system and a normal or
   high protein diet (n = 6). log‐rank (Mantel–Cox) test. b,c) G1:
   control; G2: COSlated urease with hydrogel; G3: NP with hydrogel; G4:
   NP‐laden fusogenic liposome with hydrogel; G5: NP‐laden UT‐B
   reconstituted fusogenic liposome with hydrogel; G6: FLNP with hydrogel.
   (c,e,g,i,j,m) Data are presented as mean ± s.d. *p < 0.05; **p < 0.01;
   ***p < 0.001; ****p < 0.0001 for Kaplan–Meier survival analyse,
   Student's t‐test and two‐way ANOVA test.

   Urea is a major nitrogenous waste product of protein metabolism.^[
   [145]^27 ^] The above results demonstrate that delivery of urea to
   tumor cells through UT‐B while blocking the decomposition of urea and
   increasing the accumulation of ammonium can kill tumors effectively.
   Hence, we hypothesis that high‐protein diet increases the urea
   concentration in urine, which inspires us to explore whether the
   continuous high urea stimulation could increase the antitumor efficacy
   of FLNP (Figure [146]6d). In tumor‐bearing mice fed with high‐protein
   diet, urea was highly concentrated in urine up to 1525.21 mmol L^−1
   (Figure [147]6e). While mice that fed with regular diet and water,
   urinary urea concentration was ≈684.38 mmol L^−1. The urea
   concentration in urine of mice on a high‐protein diet was ≈twofold
   higher than that of mice on a normal diet. Compared with the rapid
   tumor growth curve of the normal diet group, group fed the high‐protein
   diet showed a relatively smooth growth curve. As shown in
   Figure [148]6f–i and Figure [149]S19, Supporting Information, the
   high‐protein group led to significant tumor regression. The image and
   weight results of tumor also show that high protein diet can
   significantly enhance the therapeutic effect of FLNP on bladder cancer
   (Figure [150]6h,i). Consequently, the mice treated with FLNPs in the
   high‐protein diet group exhibited significantly prolonged survival,
   indicating that the high protein diet resulted in improved antitumor
   effects of FLNPs, largely due to the enhanced urea concentration and
   intracellular ammonia accumulation induced by synergistic anticancer
   efficacy of urease and siCPS1 (Figure [151]6j).

   To further test the effect of a high protein diet on the antitumor
   efficacy of FLNPs, we established an aggressive orthotopic bladder
   tumor model of patient‐derived xenografts (PDXs) that closely mimicked
   the tumors from cancer patients (Figure [152]6k). PDX bladder cancer
   model in NOG mice were established by using surgical specimens obtained
   from patient with bladder cancer undergoing radical cystectomy. The
   mice were then divided into three groups, followed by various
   treatments. Encouragingly, treatment with FLNPs and a high protein diet
   produced the greatest inhibition of tumor growth; the treatment had
   significantly higher inhibitory ratios than those from the PDX model
   receiving a normal diet (Figure [153]6l and Figure [154]S20a,b,
   Supporting Information). We further monitored the effects of different
   treatments on the long‐term survival time of bladder cancer‐bearing
   mice. The FLNP‐treated tumor‐bearing mice in the high protein diet
   group survived longer than those in the normal diet group
   (Figure [155]6m). These data confirm that a high protein diet
   contributes toward inhibiting tumor growth in PDX models, which is
   favorable for maximizing the therapeutic efficacies of synergistic
   ammonium metabolism treatment. To further validate the translational
   potential of our study, protein powder supplementation was performed
   preoperatively for bladder cancer patients undergoing radical
   cystectomy. Patient urine samples were collected before and after a
   high‐protein diet. As shown in Figure [156]S21, Supporting Information,
   the urea concentration of urine in the high protein diet group was much
   higher than that in the normal diet group. These results suggest that
   instilling nanoparticle‐hydrogels in combination with a high protein
   diet holds promise for the clinical treatment of bladder cancer.

3. Conclusions

   In summary, we reported an intravesical instillation
   nanoparticle‐hydrogel hybrid system for cascade metabolic modulation of
   intracellular ammonia to combat bladder cancer. We found that the
   FLNP‐mediated UT‐B immobilization dramatically induced urea
   transportation into tumor cells, and co‐delivery of urease and siCPS1
   significantly boosted ammonia accumulation in tumors inducing cell
   apoptosis. Treatment with the hybrid system exhibited superior
   antitumor effects in an orthotopic bladder tumor mouse model and a PDX
   model, respectively. Combined with high‐protein diet increased the
   production of urinary urea, lead to an augmented intracellular
   deposition of ammonia in BC cells, and ultimately an enhanced tumor
   inhibition. Together, our work established that harnessing urinary urea
   tumor‐tropistic transportation and its ammonia transformation could
   induce tumor apoptosis and may be a practical strategy for eradication
   of BC with TP53 mutation.

4. Experimental Section

Materials

   COS, hydrochloride, and HA were purchased from Shanghai Macklin
   Biochemical Co., Ltd.; DOTAP and DSPE‐PEG‐NHS were obtained from
   Shanghai Aiweite Pharmaceutical Technology Co., Ltd.; the
   2′,7′‐dichlorofluorescein diacetate (DCFH‐DA), propidium iodide (PI),
   calcein AM, JC‐1, and ATP assay kits were purchased from Solarbio Life
   Sciences. Annexin V/PI Apoptosis Detection Kit was purchased from BD
   Bioscience. The Bld‐1 peptide was purchased from GenScript
   Biotechnology Co., Ltd.; siRNA targeting CPS1 mRNA was synthesized by
   TSingke Biotechnology Co., Ltd.; and DAPI, Ki67 and TUNEL kits were
   obtained from ThermoFisher. Dil, LysoTracker and WGA lectin were
   provided by Dalian Meilun Biotech Co., Ltd and Gene Tex. The UT‐B,
   CPS1, MCL‐1, BCL‐2, p‐ERK1/2, and t‐ERK1/2 antibodies for western blot
   analysis were obtained from Abcam or Cell Signaling Technology.

Synthesis of COSlated Urease

   0.1 mg mL^−1 urease was added to protonated COS solution (0.2 g mL^−1)
   and stirred at room temperature for 30 min. Afterward, glutaraldehyde
   was added to the abovementioned mixed solution as the crosslinker.
   Finally, the COSlated urease was purified and concentrated by
   ultracentrifuge.

Synthesis of DSPE‐PEG‐NHS‐Bld‐1

   In brief, the conjugate was prepared by dissolving 5 mg of Bld‐1 and
   20 mg of DSPE‐PEG‐NHS in DMF at room temperature, and triethylamine was
   added to adjust the pH to 7–8 with slow stirring. After stirring in the
   dark for 3 days, the obtained mixture was dialyzed at 3500 Da cut off
   molecular weight in deionized water for 48 h. Finally, the obtained
   product was lyophilized and kept at −20 °C.

Construction of the Nanoporter

   For the urease and siRNA‐coloaded NP, the siRNA was condensed with the
   COSlated urease at various weight ratios (1:1, 2:1, 4:1, 8:1, and 16:1)
   at room temperature for 30 min to form urease and siRNA‐coloaded NPs.
   The formation of the NP was confirmed using a gel retardation assay on
   a 2% agarose gel.

Liposomal Coating

   DMPC, DSPE‐PEG‐Bld‐1, and DOTAP were mixed at a molar ratio of 65:3:17
   to prepare the fusogenic liposome coatings. The lipid films were
   synthetized by evaporating the solvent and hydrating with payload‐NP
   solution. Then the NP‐hydrated lipid was kept heating to 40 °C for
   10 min with stirring. The collected mixture was physically extruded
   through a 200 nm polycarbonate membrane for 20 times in total.

   To construct the FLNP, UT‐B (isolated and purified from red blood cell
   membranes) was reconstituted onto the fusogenic liposome as previously
   reported.^[ [157]^28 ^] Briefly, 1 mg UT‐B was mixed with FLNPs in 1 mL
   HEPES buffer solution. Then, the suspension was mixed to homogeneity
   and quickly frozen in liquid nitrogen for 5 min, followed with
   subsequently sonicating in a bath sonicator for 30 s. Finally, the
   suspension was centrifuged at 2000 rpm, and washed three times to
   remove the free UT‐B.

   DLS (Zetasizer ZS90, Malvern Instruments) was used to measure
   nanoparticle size and zeta potential, and JEM‐1200 EX II electron
   microscope was used to visualize the structural morphology. In short,
   5 µL samples were dropped onto the TEM grid and dried off excess
   solvent after 1 min. The experiment was replicated four times and
   averaged.

Co‐loading of Urease and Carbamoyl Phosphate Synthetase 1 in Fusogenic
Liposomalized Nanoporter Assay

   FLNP was dissociated using 1% Triton‐X100 in PBS and the mixture was
   placed in a shaker at 300 rpm min^−1 for 2 h to dissociate the FLNP and
   release of urease and siRNA. After being centrifuged at 12 000 rpm at
   4 °C for 10 min, the liposome‐free supernatant was collected. The
   urease was detected in the supernatant by using ELISA (Abcam, cat:
   ab34806). And the siRNA was detected by gel assay and Nanodrop One C
   Microvolume UV–vis spectrophotometer.

Synthesis and Gelation of Hyaluronic Acid‐DA Conjugate

   Briefly, 0.5 g HA was dissolved in 0.5 m MES buffer to form 1% HA
   solution (w/v). Then, 0.25 g EDC and 0.15 g NHS were added to HA
   solution as prepared above, and the mixture was stirred for 30 min at
   pH 5.0. Next, DOPA·HCl was added to the mixture, followed by stirring
   overnight at room temperature at pH 5.0. Unreacted DOPA·HCl was removed
   through dialysis against 1× PBS three times. The final product of the
   HA‐DA conjugate was confirmed using NMR. For hydrogel preparation,
   HA‐DA conjugate dissolved in PBS, NaIO4 at a 1:1 molar ratio to DOPA
   were mixed thoroughly to form the 2% HA‐DA hydrogel (w/v). Rheometer
   (Anton Paar MCR302) was used to detect the mechanical properties of
   HA‐DA hydrogel. The value of dynamic moduli (storage modulus G′ and
   loss modulus G″) was obtained.

Patients

   The Ethics Committee of the Qilu Hospital of Shandong University
   approved all research processes and specimens’ collection in this study
   (KYLL‐2021‐267). All the patients involved in this study provided
   informed consent. Urine from bladder cancer patients was collected on
   admission. Then, the patients were given a high‐protein diet as
   required for 2 days. After that, the urine from patients was collected
   again. Fresh bladder tumor samples were obtained through surgery for
   further experiments.

Animals

   6‐ to 8‐week‐old C57BL/6 mice and female Sprague–Dawley (SD) rats were
   obtained from the Laboratory Animal Center of Shandong University.
   6‐week‐old female Cg‐Prkdcscid Il2rg^tm1Sug/JicCrl (NOG) mice were
   obtained from Vital River Laboratory Animal Technology. All animal
   experiments were approved by the ethics committee of the Qilu Hospital
   of Shandong University (KYLL‐2020‐264).

Cell Culture

   The human bladder cancer cell line T24 was obtained from American
   Tissue Culture Collection (ATCC) and cultured in McCoy's 5A with 10%
   fetal bovine serum (FBS). The mouse bladder cancer cell line MB49 were
   purchased from Millipore and maintained in RPMI‐1640 media supplemented
   with 10% FBS.

In Vitro Cell Proliferation Assay

   To detect the cell proliferation, cells were seeded in 96‐well plates
   at a density of 3 × 10^3 cells per well. After different treatments for
   48 h, the plates were fixed with precooled trichloroacetic acid, and
   then stained with 0.4% sulforhodamine B in 1% acetic acid. The
   absorbance was determined at a wavelength of 560 nm in 10 mm Tris‐based
   solution.

Apoptosis Assay

   Annexin V/PI apoptosis detection kit was used to determine the cell
   apoptosis. In brief, cells were cultured in 6‐well plates and treated
   differently. After washing, the single‐cell suspension was collected
   and incubated with Annexin V and PI successively according to the
   instruction. Then, the apoptotic rates were analyzed by flow cytometry.

Western Blot Analysis

   After different treatment, cells were lysed in RIPA buffer supplemented
   with protease inhibitors on ice, and centrifuged at 12 000 × g to
   remove the debris. The supernatant was mixed with SDS buffer and boiled
   for 10 min. Samples were resolved in polyacrylamide gel
   electrophoresis, then incubated with indicated primary and secondary
   antibodies. The blots were visualized using detection reagents.

Oxygen Consumption Measurement

   Seahorse XFe96 Flux Analyzer was used to measure the OCR according to
   standard protocols. Briefly, cells were seeded into XFp Seahorse plates
   and treated differently for 24 h. Before measurement, the cells were
   washed with PBS and changed to seahorse base medium free of phenol red
   and bicarbonate followed by equilibration for 30 min in a 37 °C
   incubator without CO[2]. Then, the value of cellular OCR was obtained.

ATP Assay

   For the detection of intracellular ATP levels, the bioluminescent ATP
   assay kit (FLASC, Sigma) was used following the standard protocol.
   Briefly, cells were seeded in 6‐well plates along with the indicated
   treatment. For the measurement of cellular ATP levels, the cells were
   collected and resuspended in lysis reagent, and the ATP extracts were
   mixed with luciferase reagent. Then, the luminescence signals were
   measured by an IVIS Spectrum imaging system.

Confocal Laser Scanning Microscope Analysis

   For fluorescence imaging, bladder cancer cells were grown on glass
   coverslips, treated differently, and then fixed and stained with
   different fluorescent dyes as needed.

Microfilament Structure Staining

   For visualization of microfilament structures, after being washed with
   PBS, the cells were incubated with phalloidin in PBS for 2 h followed
   by DAPI staining for 1 min at room temperature. After three washes, the
   cells were mounted and analyzed using a confocal microscope for
   microfilament and nuclear morphology using a 63× objective.

In Vitro Mitochondrial Membrane Potential Staining

   The cyanine dye JC‐1 (5, 5′, 6, 6′‐tetrachloro‐1, 1′, 3,
   3′‐tetraethylbenzimi‐ dazolylcarbocyanine iodide) is a fluorescent
   probe for mitochondrial behavior, which can change from normal green
   fluorescent dye to red fluorescent aggregate when response to higher
   mitochondria membrane potential. In this study, a confocal microscope
   was used to display the alternation of mitochondria. Briefly, the cells
   were cocultured with JC‐1 at 37 °C for 15 min followed by DAPI staining
   at room temperature and then washed for imaging on a confocal
   microscope.

Urea Transporter‐B Location/Colocalization Assay

   To examine the UT‐B location on bladder cancer cells, cells were
   stained with UT‐B antibody. Cell membranes were stained with Dil, and
   nuclei were stained with DAPI. Following rinsing with PBS, the
   colocalization of the cell membrane and UT‐B was evaluated under a
   microscope.

Lysosome Staining

   Cells were incubated with LysoTracker red and UT‐B antibody (labeled by
   a fluorescent secondary antibody) for 30 min. After staining the
   nucleus with DAPI, images were captured with a confocal microscope.

Ki67 Staining

   For fluorescence imaging, the tissue sections were deparaffinized by
   xylene, dehydrated in ethanol, and boiling for antigen retrieval. After
   blocking, the sections were incubated with anti‐Ki67 antibody and
   plasma membrane probe WGA at 4 °C, followed with DAPI staining for
   nuclei. The confocal microscopy was used to capture the images.

Terminal Deoxynucleotidyl‐Transferase‐Mediated dUTP Nick‐End Labeling
Staining

   One Step TUNEL Apoptosis Assay Kit was used for DNA fragmentation and
   cell apoptosis detection. In brief, the samples were treated with TdT
   enzyme and stained, followed with DAPI staining for nuclei. Fluorescent
   images were detected by visualization with a confocal microscope.

Orthotopic Bladder Cancer Mouse Model

   The bladder cancer mouse model was established by intravesical
   injection of bladder cancer cells. Briefly, the anesthetized mice were
   inoculated with luciferin‐expressing MB49 cells into the bladder wall
   using insulin syringes.

Chemically Induced Bladder Cancer Rat Model

   The most common preclinical animal model of bladder cancer uses a
   nitrosamine compound, MNU, which causes high‐grade and invasive tumors
   of the urinary bladder. Briefly, under anesthesia, rat bladders were
   catheterized with disposable venous catheters, and then the MNU
   solution was irrigated into bladder through a sterile syringe. The rats
   were administered MNU every 2 weeks.

Patient‐Derived Xenograft Mouse Model

   A PDX bladder cancer mouse model was established using NOG mice. The
   surgical specimens from bladder cancer patients’ tumors were collected
   and divided into small pieces and then transplanted subcutaneously into
   anesthetized NOG mice. Once the tumors were grown in F0 mice, the
   xenografts were collected and further transplanted into another mouse
   for expansion. In general, F2‐ or F3‐ generation PDX tumors were used
   for further study.

Tumor‐Bearing Animal Model Treatment

   Several days after model establishment, the animals were grouped
   according to the tumor burden and treated regularly. For bladder
   perfusion, the anesthetized animals were catheterized and perfused with
   different formulations of the nanoparticle‐hydrogel hybrid system. Mice
   were sacrificed when the primary tumors reached a threshold, and the
   bladders were harvested for following experiments.

Bioluminescence Live Imaging

   The tumor burdens were measured using bioluminescence imaging with an
   IVIS Spectrum imaging system (PerkinElmer). After intraperitoneally
   injected with D‐luciferin, mice were anesthetized and detected for
   luminescence signal. The Living Image Software was used for the images
   analyzed.

Ultrasound Imaging

   Bladder tumor growth was monitored, and the antitumor effect of FLNPs
   was assessed using ultrasound imaging. Under anesthesia, the bladders
   were catheterized and distended with PBS using venous catheters and
   sterile syringes for imaging. Each animal was placed in the supine
   position on the platform, and the Vevo LAZR high‐resolution ultrasound
   system was used to acquire the images.

Hematoxylin and Eosin Staining

   At the animal experimental end point, the bladder tumors and main
   organs were collected for further analysis. After fixation, the tissues
   were embedded in paraffin, sliced into sections and histologically
   examined following hematoxylin and eosin staining.

Immunohistochemical Staining

   The samples were sectioned, deparaffinized, dehydrated and boiled in
   citrate buffer. Then immunohistochemical staining was performed using
   human polyclonal antibodies against CPS1. All images were acquired
   using a VS120 Olympus microscope (Olympus).

Urea Nitrogen Detection

   Urea is hydrolyzed by urease to produce ammonium ions and carbon
   dioxide, and ammonium ions react with phenol chromogenic agents to form
   blue substances in alkaline environments. The amount of the blue
   substance was directly proportional to the content of urea, which was
   determined by colorimetry at a wavelength of 640 nm. The concentration
   of urea nitrogen in urine was detected by a urea assay kit following
   the manufacturer's instructions.

Blood Biochemistry Analysis

   ≈1 mL of blood from each mouse was collected. The upper sera were
   collected by centrifuging at 3000 rpm for subsequent biochemical
   analysis. The serum levels of urea nitrogen (UREA), creatinine (CREA),
   aspartate aminotransferase (AST), alanine aminotransferase (ALT),
   alkaline phosphatase (ALP) and glucose (GLU) were evaluated using an
   automatic chemistry analyzer (BS‐240VET).

Statistics Analysis

   Data analyses in this study were carried out with GraphPad Prism 8.0.
   The results are presented as the mean ± standard deviation (s.d.) and
   based on experiments performed at least in triplicate, n = 3–50
   depending on the condition being presented. Difference between two
   groups were analyzed using Student's t‐test. One‐way ANOVA test was
   used to analyze differences among multiple groups. Two‐way ANOVA test
   was used to compare differences between two or more groups with
   different time points. Kaplan–Meier survival analyses were performed
   with the log‐rank test and used for survival data of every group.
   Figure legends specify the statistical strategies and data
   presentation. p < 0.05 was considered statistically significant
   (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001).

Conflict of Interest

   X.J., W.J. and C.C. are inventors on patent application submitted by
   Shandong University that covers the following content: Metabolic
   Modulation of Intracellular Ammonia Eradicates Bladder Carcinoma in a
   Murine Model of Patient‐Derived Xenograft. The other authors declare
   that they have no other competing interests.

Author Contributions

   W.J. and C.C. contributed equally to this work. X.J., B.S., W.J., and
   C.C. conceived the project and designed the experiments. W.J., C.C.,
   G.W., M.H., S.C., M.A., X.J., C.S., P.S., and Z.Y. performed the
   experiments. W.J., C.C., G.W., M.H., B.S., and X.J. analyzed and
   interpreted the data in this study. W.J., C.C., B.S., and X.J. wrote
   the manuscript draft.

Supporting information

   Supporting Information
   [158]Click here for additional data file.^ (1.6MB, pdf)

Acknowledgements