Synergistically Enhanced Mucoadhesive and Penetrable Polypeptide Nanogel for Efficient Drug Delivery to Orthotopic Bladder Cancer

Intravesical chemotherapy has been recommended after the gold standard of transurethral resection of the bladder tumor to prevent bladder cancer (BC) from local recurrence in the clinic. However, due to rapid urine excretion and barrier protection of the bladder wall, the clinical performances of chemotherapeutic drugs are severely compromised. In the present work, a smart positively charged disulfide-crosslinked nanogel of oligoarginine-poly(ethylene glycol)–poly(L-phenylalanine-co-L-cystine) (R9-PEG–P(LP-co-LC)) was prepared to prolong the retention period and enhance the penetration capability of chemotherapeutic agent toward the bladder wall. PEG significantly improved the aqueous dispersibility of the 10-hydroxycamptothecin (HCPT)-loaded R9-PEG–P(LP-co-LC) (i.e., R9NG/HCPT) and enhanced the mucoadhesive capability by the nonspecific interaction between PEG chain and the bladder mucosa accompanied with the electrostatic interaction between the cationic R9 and negatively charged bladder mucosa. Besides, R9, as a cell-penetrating peptide, efficiently penetrated through the cell membrane and delivered carried cargo. The disulfide bond endowed the selective release behavior of HCPT triggered by the intracellular reductive microenvironment. As an advanced chemotherapeutic nanoformulation, the smart R9NG/HCPT demonstrated superior cytotoxicity against human BC 5637 cells in vitro and remarkably enhanced tumor suppression activity toward orthotopic BC models of mouse and rat in vivo, indicating its great potential in the clinical intravesical BC chemotherapy.


Introduction
Costs of curing bladder cancer (BC) are the highest on a perpatient basis because of the elevated rate of recurrence and multiple therapeutic interventions [1]. Over 75% of BC cases are initially diagnosed with non-muscle-invasive bladder cancer (NMIBC) [2]. The recommendation for NMIBC patients after transurethral resection of the bladder tumor is immediate intravesical chemotherapy, which has been proven to help prevent cancer from recurrence [3]. However, the clinical performances of chemotherapeutic drugs are severely compromised by rapid urine excretion and barrier protection of the bladder wall. Approximately 75% of patients will experience tumor relapse or progression within five years, which significantly affects their survival [4]. The five-year survival rate of BC patients has not been improved in the past three decades [5].
Nanomedicine, as revolutionary manufacturing technology, is expected to provide great opportunities to upregulate the exposure periods of drugs within the bladder. Currently, the leading materials, such as poly(ethylene glycol) (PEG) [6], chitosan [7], and liposomes [8], have been developed to improve the mucoadhesiveness or penetrability. PEG seems to be the most promising mucoadhesive polymer due to the nonspecific interaction between PEG chain and mucous membrane [9]. Moreover, PEG possesses the capability of improving the water solubility of hydrophobic drugs [10] and is well known for its reduced toxicity and low immunogenicity [11]. Cell-penetrating peptides (CPPs) are potentially less cytotoxic and possess the capacity of permeating through cell membrane to transport carried cargoes [12,13]. The combination of CPPs with anti-tumor drugs has been considered an incredibly attractive concept to improve therapeutic efficacy [14]. Among them, the arginine-rich CPPs have been demonstrated to be highly efficient in terms of cell uptake, especially the oligoarginine of nine arginine residues (R 9 ) [15,16].
Accordingly, an innovative nanogel of R 9 -poly(ethylene glycol)-poly(L-phenylalanine-co-L-cystine) (R 9 -PEG-P(LP-co-LC)) was successfully synthesized to improve the mucoadhesiveness and penetrability of chemotherapeutical drugs (Scheme 1A). 10-Hydroxycamptothecin (HCPT) was applied as a model drug and encapsulated into the core of R 9 -PEG-P(LP-co-LC) by facile diffusion. The obtained drug-loaded nanogel was referred as R 9 NG/HCPT. The morphology of R 9 NG/HCPT is similar to that of an octopus with spears in its hands, as shown in Scheme 1B. The rinsing of R 9 NG/HCPT caused by urine excretion was resisted by the presence of PEG chains, which were like arms of an octopus, and R 9 , which enhanced the mucoadhesion of R 9 NG/HCPT through the nonspecific and electrostatic interactions, respectively. The powerful penetrating capacity of R 9 overcame barrier protection of the bladder wall. R 9 was the weapon in the hands of octopus to break through obstacles. The drug-loaded core of polypeptide nanogel was much like the head of the octopus, which released the cargo triggered by the high concentration of glutathione (GSH) in BC cells [17].
The nanogel composed of PEG and polypeptide, but without R 9 , was synthesized as a negative control (i.e., NG/HCPT). In order to better understand the anti-tumor efficacy of R 9 NG/HCPT in vivo, different orthotopic BC models of mouse and rat were constructed. The orthotopic BC model in mouse was constructed by instilling MB49 cells into the bladder that was pretreated with PLL solution [19]. The orthotopic BC model in rat was successfully established by intravesical instillation of N-methyl-N-nitrosourea (MNU). MNU played a vital role in the process of carcinogenesis, which was histologically indistinguishable from the human tumor [20]. The physicochemical properties, intracellular release profiles, and cytotoxicity of R 9 NG/HCPT in vitro and mucoadhesiveness, permeability, biodistribution, antitumor activity, and systemic toxicity in vivo were systematically demonstrated. Encouragingly, R 9 NG/HCPT possessed unique physicochemical properties to improve the chemotherapy efficacy of BC both in vitro and in vivo, as a potential benefit of strong urothelium adhesive property, high penetrating capacity, and selective drug release behavior.

Results and Discussion
2.1. Preparation and Characterizations of Empty and HCPT-Loaded R 9 -PEG-P(LP-co-LC). aPEG-P(LP-co-LC) was syn-thesized by the ROP of LP NCA and LC NCA with the amino-terminated allyloxy poly(ethylene glycol) (aPEG-NH 2 ) as a macroinitiator. R 9 -PEG-P(LP-co-LC) was synthesized by the decoration of R 9 at the end of PEG by sequential deprotection, amidation, and Michael addition reactions, as depicted in Scheme 1A. The chemical structures of intermediates and the desired product R 9 -PEG-P(LP-co-LC) were confirmed by the proton nuclear magnetic resonance ( 1 H NMR) spectra. The characteristic resonance signals at 7.19 ppm (e; -CH 2 C 6 H 5 , 5H) and 3.46 ppm (d; -CH 2 C 6 H 5 , 2H) were assigned to the protons in the L-phenylalanine unit. The peak at 2.42 ppm (g; -SCH 2 -, 2H) was ascribed to protons of the L-cystine unit. The signal at 4.55 ppm (c+f; -COCH(CH 2 )NH-, 1H) was attributed to the protons of peptide backbone both in the polypeptide and R 9 . The strong and broad signal peak of b at 3.82 ppm (b; -OCH 2 CH 2 -, 4H) was assigned to the methylene protons of PEG. The signal at 2.42 ppm (a; -SCH 2 CH 2 CH 2 -, 2H) was produced by the thiol-ene click reaction between allyloxy and thiol groups. The peaks at 5.28 and 5.82 ppm (a and b; CH 2 =CH-, 2H and 1H, respectively) were attributed to the allyloxy protons of aPEG-P(LP-co-LC), as shown in Supplementary Figure S1. The signal disappearance of the allyloxy protons (i) and the appearance of the t-butyloxycarbonyl (t-Boc) resonance signal at 1.52 ppm (j; (CH 3 ) 3 C-, 9H) indicated the successful synthesis of t-Boc-NH-PEG-P(LP-co-LC), as depicted in Supplementary Figure S2 . The disappearance of t-Boc peak (j) demonstrated the successful synthesis of NH 2 -PEG-P(LP-co-LC), as shown in Supplementary Figure S3. The maleimide-terminated polymer was produced by a simple amidation reaction, which was confirmed by the signal at  Figure S4A , and the signal disappearance of two vinyl protons on maleimide at 6.48 ppm (h) both confirmed the successful synthesis of R 9 -PEG-P(LP-co-LC). The nanogel of Mal-PEG-P(LP-co-LC) was synthesized as a negative control. The molar ratio of LP and LC in R 9 -PEG-P(LP-co-LC) was calculated as 9:5 according to the content of carbon (C), hydrogen (H), nitrogen (N), and sulfur (S) determined by elemental analysis.
( Figure 1(c)). The results demonstrated the reductionresponsive properties of NG/HCPT and R 9 NG/HCPT, which were critical for the selective delivery of drugs to cancer cells. The zeta potential of NG/HCPT was −17:3 ± 0:8 mV, while that of R 9 NG/HCPT was 26:9 ± 1:2 mV. The positive charge of R 9 NG/HCPT further demonstrated the successful conjugation of R 9 onto the surface of the nanogel, which could enhance the interaction between R 9 NG/HCPT and negatively charged cell membrane and subsequently upregulated cell internalization [21]. The release behavior of HCPT from R 9 NG/HCPT was examined in PBS at pH 7.4 with 0.1% (W/V) Tween-80 containing different concentrations of DTT. As shown in Figure 1(d), free HCPT displayed a rapid diffusion from the dialysis membrane within 12 h. In contrast, R 9 NG/HCPT released only 38.2% of HCPT throughout 72 h in PBS without DTT, demonstrating excellent stability of R 9 NG/HCPT under physiological conditions. In addition, the release rate of HCPT from R 9 NG/HCPT was remarkably accelerated due to the increased concentration of DTT. The cumulative release amount of HCPT was 62.9% and 94.5% over a period of 72 h with the concentration of DTT ranging from 5.0 to 10.0 mM, respectively. The results demonstrated the excellent physiological stability of R 9 NG/HCPT. However, the disulfide bond could be rapidly broken in the reductive microenvironment, resulting in the selective HCPT release in the cancer cells.

Intracellular
Release and Cytotoxicity of R 9 NG/HCPT. The intracellular release behavior of R 9 NG/HCPT was qualitatively evaluated by confocal laser scanning microscopy (CLSM). Figure 2(a) presents the CLSM images of human BC 5637 cells after co-incubation with free HCPT, NG/HCPT, or R 9 NG/HCPT with equivalent HCPT concentration at 1.25 μg mL −1 . The cells depicted slightly weaker HCPT fluorescence after co-incubation with R 9 NG/HCPT for 2 h as compared with that of free HCPT. At the same point, the fluorescence intensity of negatively charged NG/HCPT was the weakest. Although it was not as simple as small-molecule drugs entering cells by diffusion, the results demonstrated that R 9 NG/HCPT could be internalized by human BC 5637 cells effectively [22]. Furthermore, after the co-incubation period increased up to 6 h, the cells treated with R 9 NG/HCPT exhibited more robust HCPT fluorescence than that of free HCPT or NG/HCPT ( Figure 2(a)). The significantly enhanced HCPT fluorescence mainly benefited from the electrostatic interaction between the positively charged R 9 NG/HCPT and the negatively charged cell membrane [23]. The conjugated R 9 facilitated the cell uptake of R 9 NG/HCPT through the CPP-mediated cell entry [24]. In addition, the high concentration of GSH in human BC 5637 cells led to the rupture of the disulfide bond and accelerated the HCPT release from R 9 NG/HCPT [25]. The cell uptake and subsequent release profiles were investigated quantitatively according to the protocol described by Wei et al. [26]. Compared with free HCPT, the cells cocultured with R 9 NG/HCPT displayed a slightly lower content of HCPT at 2 h. The accumulation of HCPT in cells cocultured with NG/HCPT was the lowest at the same time point. With the prolongation of co-incubation time, the endocytosis of R 9 NG/HCPT increased significantly, and a relatively higher content of HCPT was maintained as compared to that of free HCPT or NG/HCPT from the second time point. Conversely, the increased HCPT concentration in the cells of the R 9 NG/HCPT group was about 1.9 times compared with that in the cells treated with free HCPT (Figure 2(b)), while it was about 1.5 times as much as that in the NG/HCPT group at 6 h. The results were consistent with the representative finding by CLSM (Figure 2(a)).
To evaluate the potential cytotoxicity of R 9 NG/HCPT against human BC 5637 cells, a standard methyl thiazolyl tetrazolium (MTT) assay was performed. The cytotoxicity of free HCPT, NG/HCPT, and R 9 NG/HCPT was enhanced in a concentration-dependent manner (Figure 2(c)). Furthermore, the cytotoxicity of R 9 NG/HCPT surpassed those of free HCPT and NG/HCPT against human BC 5637 cells. The superior cytotoxic effect of R 9 NG/HCPT probably benefited from the weak positive surface, R 9 , and the reduction-responsive property of R 9 NG/HCPT, which led to the promoted cell uptake and accelerated intracellular HCPT release in human BC 5637 cells. Importantly, the half-maximal inhibitory concentration (IC 50 ) of R 9 NG/HCPT was 2:4 ± 0:1 μg mL −1 , which was lower in comparison with those of free HCPT (i.e., 8:7 ± 0:8 μg mL −1 ) and NG/HCPT (i.e., 4:7 ± 0:2 μg mL −1 ). The lowest IC 50 indicated the potent cytotoxicity of R 9 NG/HCPT quantitatively, suggesting the superiority of R 9 NG/HCPT for intravesical chemotherapy of BC in vivo.
The R 9 NG/HCPT-induced apoptosis of human BC 5637 cells was measured by flow cytometry (FCM) analysis. As shown in Figure 2(d), the 24 h exposure to three different formulations of HCPT at an equivalent HCPT dosage of 0.1 μg mL −1 resulted in a noticeable difference in the percentage of apoptotic cells. R 9 NG/HCPT contributed most to an abundance of apoptotic cells. The total percentage of apoptotic cells was 22.5%, which was constituted with 16.5% early apoptotic cells and 6.0% late apoptotic cells. In contrast, the total percentage of cells displaying apoptosis was reduced to 12.2% and 18.7% after co-incubation with free HCPT and NG/HCPT, respectively. The results were consistent with the cytotoxicity of R 9 NG/HCPT described above. When the co-incubation time was prolonged to 48 h, the differences among free HCPT, NG/HCPT, and R 9 NG/HCPT groups varied greatly, as shown in Supplementary Figure S5. The total percentage of cells displaying apoptosis was increased to 38.9%, 53.5%, or 63.0% after co-incubation with free HCPT, NG/HCPT, or R 9 NG/HCPT, respectively. This phenomenon indicated that different formulations induced the apoptosis of human BC 5637 cells in a time-dependent manner. The increased apoptotic activity induced by R 9 NG/HCPT benefited from the high intracellular concentration of HCPT, which was internalized through CPP-mediated endocytosis and triggered to release by GSH rapidly.
2.3. Mucoadhesiveness, Permeability, and Biodistribution of R 9 NG/HCPT. CLSM was employed to demonstrate the mucoadhesiveness and permeability of R 9 NG/HCPT. The male Sprague-Dawley (SD) rats with BC were anesthetized and then administrated with different HCPT formulations at a consistent HCPT dosage of 6.0 mg per kg body weight (mg (kg BW) −1 ) through bladder irrigation. At different indicated time points (i.e., 0.5, 2, 6, 12, 24, and 48 h), the bladders were removed and sectioned.
As depicted in Figure 3(a), the fluorescence signals of free HCPT and R 9 NG/HCPT were observed in the bladder sections at the first few time points. The fluorescence intensity of R 9 NG/HCPT was stronger than that of free HCPT. The highest fluorescence signal was observed in the bladder treated with R 9 NG/HCPT at 0.5 h. Not surprisingly, the fluorescence intensity of R 9 NG/HCPT decreased over time. However, it remained at a relatively high concentration even at 12 h, indicating the high mucoadhesion of R 9 NG/ HCPT toward the urothelial surface. On the contrary, the fluorescence signal in the bladder exposed to free HCPT decreased rapidly, which was attributed to the continuous excretion of urine. This phenomenon demonstrated the outstanding mucoadhesiveness of R 9 NG/HCPT. In order to quantitatively prove the mucoadhesiveness of R 9 NG/HCPT, the average optical density of HCPT was measured. As verified in Figure 3(b), the fluorescence intensity of R 9 NG/HCPT was 1.6 times higher as compared with that of free HCPT at 0.5 h. Inspiringly, the fluorescence intensity increased from 2.2 to 4.8 times when the retention time was prolonged from 2 to 6 h. The outstanding mucoadhesiveness of R 9 NG/HCPT benefited from the nonspecific interaction between PEG chain and the bladder mucosa and electrostatic interaction between terminal cationic R 9 and negatively charged surface  of the bladder mucosa, which retarded the wash-off of R 9 NG/HCPT by urine [27,28]. The permeability of R 9 NG/HCPT was verified in Figure 4(a). Within half an hour, a much stronger fluorescence signal was observed in the full-thickness bladder wall treated with free HCPT and especially confined to the mucous membrane. At the same time, R 9 NG/HCPT displayed a shallow penetration within the bladder wall. However, R 9 NG/HCPT gradually penetrated through the full-thickness bladder over time. It also maintained a relatively stronger HCPT fluorescence than free HCPT for an extended period. The results indicated the exceptional permeability and accumulation of R 9 NG/HCPT in the bladder. For detailed analysis of the permeability of R 9 NG/HCPT, the HCPT fluorescence signals were normalized to the maximum value. As shown in Figure 4(b), the fluorescence intensity of R 9 NG/HCPT gradually spread throughout the whole bladder tissue over time and remained a comparatively higher HCPT content. The phenomenon was mainly due to the strong penetration ability of R 9 with a positive surface charge [29,30]. In addition, the advanced mucoadhesiveness of R 9 NG/HCPT further promoted its permeability. An excellent chemotherapeutic effect of R 9 NG/HCPT was predictable based on superior permeability, which was vital for the treatment of BC.
The bladder accumulation of R 9 NG/HCPT was verified by high-performance liquid chromatography (HPLC). Male SD rats were administrated with intravesical instillation of different HCPT solutions, and the experimental rats were euthanized at 6 h. Subsequently, the bladders and other major organs, such as the heart, liver, spleen, lung, and kidney, were excised and homogenized. Then, the samples were extracted with the organic phase of methanol and acetonitrile (1/1, V/V). 20.0 μL of mixture suspension collected by centrifugation was analyzed for the determination of drug content. HCPT, released from R 9 NG/HCPT, was mainly accumulated in the bladder, which was 2.2 times higher as compared with free HCPT ( Figure S6, Supporting Information). In addition, the other organs provided extremely low drug content. The results confirmed the advanced mucoadhesiveness of R 9 NG/ HCPT and predicted an improved anti-tumor efficacy of R 9 NG/HCPT with reduced toxicity on normal organs.

In Vivo
Anti-Tumor Efficacy of R 9 NG/HCPT. The anti-tumor efficacy of R 9 NG/HCPT was assessed on the orthotopic BC in C57Bl/6 mice. Briefly, PLL (100.0 μL of 0.1 mg mL −1 for 20 min) was used to disrupt the glycosaminoglycan layer. 50.0 μL of MB49 cell suspension, including 1:0 × 10 5 cells, was slowly injected into the bladder of C57Bl/6 mouse, and the cells were allowed to dwell in the bladder for 50 min [31]. The tumor-bearing mice were randomly divided into four groups: PBS as a control, free HCPT, NG/HCPT, and R 9 NG/HCPT (n = 6). Different HCPT formulations were administered weekly by catheterization at a dosage of 6.0 mg (kg BW) −1 for a total of four treatments. To define the development of BC, especially the tumor size, cystography was performed. Polypoid filling defects and irregular surface were manifested in cystography after treat-ment, which was depicted in Figure 5(a). The most extensive space-occupying lesion was identified in the bladder exposed to PBS, while there was no significant abnormality recognized in the bladder administrated with R 9 NG/HCPT based on the smooth margin. The superior anti-tumor efficacy of R 9 NG/HCPT was intuitionally approved by cystography. The result was mostly attributed to the presence of PEG and R 9 in R 9 NG/HCPT, which could eventually result in a remarkably enhanced mucoadhesiveness and permeability to improve the anti-tumor efficacy synergistically. The suboptimal BC growth inhibition of free HCPT was caused by rapid drug excretion from urine.
The anti-tumor efficacy of R 9 NG/HCPT was further assessed on another orthotopic BC in male SD rats. The orthotopic BC was established by four doses of MNU [32]. Histopathological analysis was performed to demonstrate the successful induction of BC after the intravesical instillation of MNU for eight weeks. As verified in Supplementary Figure S7A, the abundance of cancer cells was observed with apparent architectural disarray and nuclear atypia, indicating the occurrence of BC. The BC-bearing rats were randomly divided into three groups: PBS as a control, free HCPT, and R 9 NG/HCPT (n = 8). The equivalent HCPT dosage of 6.0 mg (kg BW) −1 was given by intravesical instillation weekly for a total of six treatments. Cystography was performed to define the development of BC. As pointed in Supplementary Figure S7B, the results were consistent with those of cystography in mice ( Figure 5(a)).
The bladders were harvested at the end of the treatments to assess the external surface, which was shown in Figure 5(b) and Figure S7B, Supporting Information. A rigid-walled bladder with the largest volume was detected in the PBS group, suggesting that exuberant growth of the tumor had infiltrated the muscular layer of the bladder wall. In contrast, a smooth-surfaced bladder with standard size and general shape occurred in the R 9 NG/HCPT group. The bladders exposed to free HCPT or NG/HCPT were generally less rigid, suggesting that free HCPT and NG/ HCPT were effective but less effective than R 9 NG/HCPT. To be better informed of the growth of the tumor, the bladders were opened up, and the internal surface was displayed ( Figure 5(b) and Figure S7B, Supporting Information). Multiple neoplasms with a "fish flesh" soft tan appearance were found in the bladder cavity of the control group, while there was no obvious abnormality in the bladder treated with R 9 NG/HCPT. The bladder observation intuitively confirmed that R 9 NG/HCPT possessed an excellent anti-tumor efficacy on BC both in mouse and rat.
In order to evaluate the systemic toxicity of R 9 NG/HCPT, the body weights of the experimental animals were monitored once a week and shown in Figure 5(c) and Figure S7C, Supporting Information. The experimental animals ad-ministrated with R 9 NG/HCPT showed a stable increase in body weight during the treatment, indicating negligible systemic toxicity of R 9 NG/HCPT. Another essential pre-dictor of efficacies and toxicity of various drug formulations was the survival rate, which was shown in Figure 5(d) and Figure S7D , Supporting Information [33]. Different HCPT formulations extended the overall survivals of experimental animals in comparison with the control group. Intravesical instillation of R 9 NG/HCPT has shown an especially sig-nificant survival advantage in the murine orthotopic BC model. These encouraging results benefited from the excellent biochemical properties of R 9 NG/HCPT. R 9 NG/HCPT possessed improved mucoadhesiveness and permea-bility and could selectively release HCPT within cancer cells. All these results suggested that the tumor growth inhibition effect was enhanced, while the systemic toxicity was decreased, which confirmed the reliable biological safety of R 9 NG/HCPT. To further accurately evaluate the anti-tumor activity of R 9 NG/HCPT, the histopathological examination of the bladders was performed after the completion of intravesical chemotherapy. For hematoxylin and eosin (H&E) staining (Figure 6(a) and Figure S8A, Supporting Information), large amounts of cancer cells with spherical or spindle nuclei and prominent atypia were observed in the bladder section of the PBS group. Fortunately, the tumor cells showed karyopyknosis, chromatic agglutination, apoptotic body, and ill-defined morphology in the bladder administrated with R 9 NG/HCPT, indicating the predominant contribution of R 9 NG/HCPT in tumor suppression.
Moreover, the detection of apoptotic DNA fragments was performed using the terminal deoxyribonucleotide transferase (TdT)-mediated biotin-16-dUTP nick-end labeling   (TUNEL) method. As depicted in Figure 6(b), the red fluorescence of Cy3 was observed in all the bladder sections. In particular, the bladder treated with R 9 NG/HCPT showed a larger area of flake-like red fluorescence as compared to those of the other three groups.
Caspase-3, a crucial mediator of programmed cell apoptosis, is located downstream of apoptotic response [34]. To further analyze the apoptosis mediated by R 9 NG/HCPT, immunofluorescence staining for caspase-3 was performed. The expression of caspase-3 in mouse and rat bladder tissues was depicted in Figure 6(c) and Figure S8B, Supporting Information, respectively. The level of caspase-3 was much higher in the section administrated with R 9 NG/HCPT as compared to that administrated with free HCPT or NG/ HCPT. A feeble positive signal was observed in the bladder section of control group. The phenomenon indicated that a large number of cells were undergoing apoptosis in the R 9 NG/HCPT group.
Ki-67, a cell proliferative marker, is expressed from prophase to anaphase [35]. To confirm the cell proliferation status after treatment with R 9 NG/HCPT, the expression of Ki-67 in different experimental animals was monitored by immunofluorescence, as shown in Figure 6(c) and Figure S8B, Supporting Information. The strongest positive signal was recorded in the control group. The proportion of Ki-67-positive cells diminished significantly due to the use of different HCPT formulations, especially R 9 NG/HCPT, which demonstrated that the growth of BC cells was obviously suppressed by R 9 NG/HCPT.
To further confirm the expression of caspase-3 and Ki-67, the quantitative analysis was performed by a similar method as mucoadhesiveness mentioned above. The quantitative data described the order of caspase-3 expression as R 9 NG/HCPT > NG/HCPT > free HCPT > PBS as a control. More importantly, the expression of caspase-3 in the R 9 NG/HCPT group was 7.0 ( Figure 6(d), BC in mouse) and 5.6 ( Figure S8C, BC in rat) times higher as compared to that of the free HCPT group, which demonstrated that a large number of cells were undergoing apoptosis in the R 9 NG/ HCPT group. The quantitative data of Ki-67 demonstrated that the Ki-67 signal was sorted as PBS as a control > free HCPT > NG/HCPT > R 9 NG/HCPT, demonstrating a signifi-cant proliferation inhibition of tumor cells induced by R 9 NG/HCPT (Figure 6(e) and Figure S8D, Supporting Information). All these results indicated that R 9 NG/HCPT possessed unique advantages for BC chemotherapy.

Conclusion
In summary, a cationic R 9 -decorated disulfide-crosslinked R 9 NG/HCPT with enhanced mucoadhesion and penetration toward the bladder mucosa was designed and developed for intravesical chemotherapy. PEG and R 9 endowed R 9 NG/ HCPT with upregulated mucoadhesiveness by the nonspecific and electrostatic interactions, respectively, and R 9 further promoted the permeability of laden nanogel within the bladder wall. The advanced properties of R 9 NG/HCPT overcame the disadvantages of clinical chemotherapeutic drugs, which will bring unpredictable benefits to BC patients and increase overall survival. More importantly, the reductionresponsive property enabled R 9 NG/HCPT to release the cargo specifically in cancer cells triggered by the intracellular reductive microenvironment. This means that R 9 NG/HCPT may achieve the same intracellular drug concentration as clinical chemotherapeutic agents, but the administered quantity of R 9 NG/HCPT will decrease. The property avoided the potential risk of drug resistance due to high-dose chemotherapy, which was fundamentally critical for patients with BC. The extremely outstanding anti-tumor efficacy and the negligible systemic toxicity pave the way of R 9 NG/HCPT for further progress in the clinical applications.

Data Availability
All data are available in the manuscript and supplementary materials, or from the authors.

Conflicts of Interest
The authors declare no conflict of interest.