An ATF24 peptide-functionalized β-elemene-nanostructured lipid carrier combined with cisplatin for bladder cancer treatment

Objective: In this study, we aimed to develop an amino-terminal fragment (ATF) peptide-targeted liposome carrying β-elemene (ATF24-PEG-Lipo-β-E) for targeted delivery into urokinase plasminogen activator receptor-overexpressing bladder cancer cells combined with cisplatin (DDP) for bladder cancer treatment. Methods: The liposomes were prepared by ethanol injection and high-pressure microjet homogenization. The liposomes were characterized, and the drug content, entrapment efficiency, and in vitro release were studied. The targeting efficiency was investigated using confocal microscopy, ultra-fast liquid chromatography, and an orthotopic bladder cancer model. The effects of ATF24-PEG-Lipo-β-E combined with DDP on cell viability and proliferation were evaluated by a Cell Counting Kit-8 (CCK-8) assay, a colony formation assay, and cell apoptosis and cell cycle analyses. The anticancer effects were evaluated in a KU-19-19 bladder cancer xenograft model. Results: ATF24-PEG-Lipo-β-E had small and uniform sizes (˜79 nm), high drug loading capacity (˜5.24 mg/mL), high entrapment efficiency (98.37 ± 0.95%), and exhibited sustained drug release behavior. ATF24-PEG-Lipo-β-E had better targeting efficiency and higher cytotoxicity than polyethylene glycol (PEG)ylated β-elemene liposomes (PEG-Lipo-β-E). DDP, combined with ATF24-PEG-Lipo-β-E, exerted a synergistic effect on cellular apoptosis and cell arrest at the G2/M phase, and these effects were dependent on the caspase-dependent pathway and Cdc25C/Cdc2/cyclin B1 pathways. Furthermore, the in vivo antitumor activity showed that the targeted liposomes effectively inhibited the growth of tumors, using the combined strategy. Conclusions: The present study provided an effective strategy for the targeted delivery of β-elemene (β-E) to bladder cancer, and a combined strategy for bladder cancer treatment.


Introduction
Cancer is a major cause of morbidity and mortality worldwide, regardless of the level of economic development. Worldwide, bladder cancer is the sixth most common cancer among males 1 . In China, bladder cancer is one the most common malignant tumors of the urinary system among males 2,3 . For patients with non-muscle-invasive bladder cancer (NMIBC), radical cystectomy with lymphadenectomy are the standard treatments 4 . In contrast to NMIBC, muscle-invasive bladder cancer (MIBC) is a poly-phase cancer that may metastasize to the prostate, vagina, or bowel, and has a much lower 5-year survival rate 5 . Thus, a better understanding of the molecular mechanisms responsible for the invasive behavior of bladder cancer is important.
The system consists of urokinase-type plasminogen activator (uPA)/urokinase-type plasminogen activator receptor (uPAR), which plays a critical role in fibrinolysis of the extracellular matrix, and facilitates tumor cell migration and invasion 6 . High uPAR expression has been detected in tumor tissues in patients with high metastatic cancer and a poor prognosis. In contrast, strategies that block uPAR expression decrease the invasive capacity of tumor cells [7][8][9] . Hau et al. 10 further identified increased uPAR expression in 94% of invasive human bladder cancers and in 54%-71% of noninvasive bladder cancers. Agents that inhibit uPAR expression may therefore be useful for developing antitumor therapies.
In recent years, drug delivery systems using drug-loaded active targeting liposomes have attracted attention, with the aim of increasing the number of liposomes internalized into tumor cells 11 . Liposome targeting of only tumor cell surface targets can only be delivered into the tumor interstitial space via "leaky" tumor vasculatures mediated by the enhanced permeability and retention (EPR) effect 12,13 . Molecular targets allowing for active targeting to the tumor endothelial cell layer, tumor stroma, and tumor cells can facilitate the crossing of the endothelial layer by drug-loaded liposomes, thus increasing the efficiency of liposomes in tumor cells 14 . The uPAR is primarily expressed by myofibroblasts and macrophages in the tumor-associated stroma and some tumor cells 15,16 . The high level of uPAR expression in MIBC tissues supports the development of targeted liposome drug carriers for the effective treatment of MIBC. Studies have shown that an amino-terminal fragment (ATF) peptide can compete with uPA for the binding of uPAR at the surface of both tumor and endothelial cells, resulting in the inhibition of tumor growth and angiogenesis. For example, the delivery of uPAR-targeted iron oxide nanoparticles carrying gemcitabine or cisplatin (DDP) resulted in significant growth inhibition in pancreatic tumors [17][18][19] .
β-Elemene (β-E) is a broad-spectrum antitumor drug extracted from the traditional Chinese medicine, Curcuma wenyujin Y. H. Chen et C. Ling. To date, an elemene emulsion injection and an oral emulsion have been developed and used to treat various cancers for more than 20 years 20 . Studies have shown that these emulsions can be used to enhance the efficacy and reduce the toxicity of chemoradiotherapy, and reverse drug resistance without significant side effects [21][22][23] . To overcome the drawback of low delivery efficiency of conventional elemene emulsions, we developed a novel ATF 24 peptide-targeted liposome carrying β-E (ATF 24 -PEG-Lipo-β-E) for targeted delivery into uPAR-expressing tumors and stromal cells, combined with DDP for bladder cancer treatment. The proposed mechanism of uPAR-targeted β-E liposomes combined with DDP for the treatment of bladder cancer is shown in Figure 1.

Materials
The following reagents were obtained: ATF 24

Preparation of ATF 24 -PEG-Lipo-β-E
Ethanol injection and high-pressure microjet homogenization were used to prepare the polyethylene glycol (PEG) ylated liposomes and ATF 24 modified liposomes. Specifically, soybean lecithin (2.5 g), cholesterol (0.1 g), and DSPE-PEG 2000 (0.3 g) were dissolved in 2 mL of 95% ethanol in an 80 °C water bath. The oil phase was slowly added to 100 mL of water containing 10 mM L-histidine (pH 6.5) at 60 °C, followed by stirring using an Ultra-Turrax T18 high-speed blender (IKA, Staufen, Germany) at 13,700 × g for 60 min. The solution was passed through an LM20 microfluidizer (Microfluidics, Westwood, MA, USA) at a pressure of 15,000 psi for 3 cycles, and then the PEGylated liposomes were formed 25 . The ATF 24 -modified liposomes were prepared with 0.3 g of DSPE-PEG 2000 in blank PEGylated liposomes (PEG-Lipo) being replaced by 0.2 g of DSPE-PEG 2000 -ATF 24 and 0.1 g of DSPE-PEG 2000 in a similar manner. The DiDlabeled and β-E-loaded liposomes were prepared by adding 5 mg of DiD and 0.5 g β-E, respectively, to the original formula.

Characterization of ATF 24 -PEG-Lipo-β-E
The morphology of ATF 24 -PEG-Lipo-β-E was determined using transmission electron microscopy (TEM) (HT-7700; Hitachi, Tokyo, Japan). The mean particle size, zeta potential, and polydispersity index (PDI) of the liposome droplets were determined using a Zetasizer Nano ZS (Malvern Instruments, Malvern, UK) 26 . The pH of the samples was measured using a pH meter at 20 ± 1 °C (PB-10; Sartorius, Goettingen, Germany). The drug content was detected by ultra-fast liquid chromatography (UFLC) (Shimadzu, Kyoto, Japan). The entrapment efficiency (EE) of β-E in the ATF 24 -modified liposomes was determined by the liquid surface method 27 .

In vitro release of β-E from the ATF 24 -modified liposomes
Measurement of the in vitro release of free β-E, PEG-Lipo-β-E, and ATF 24 -PEG-Lipo-β-E was conducted by using the dialysis bag method. Briefly, 5 mL of β-E, PEG-Lipo-β-E or ATF 24 -PEG-Lipo-β-E (5 mg/mL) was placed in a dialysis membrane (MW cutoff: 300,000 Da), ligated, immersed in the 75% ethanol release media (100 mL), and stirred using a magnetic force at 37 °C. One mL of the sample was removed at predetermined time intervals of 0.5, 1, 2, 4, 6, and 8 h and diluted to 10 mL. The supernatants were filtered through 0.22 µm microporous membranes and subjected to UFLC analysis for the determination of β-E 25 .

Cellular uptake
For qualitative analysis, 2 × 10 5 KU-19-19 cells were seeded on round cell slides in 12-well plates and cultured for 24 h. The cells were incubated with the DiD-labeled liposomes (5 µg/mL) for another 4 h and washed twice with ice-cold phosphate-buffered saline (PBS). Then, the cells were fixed with 4% paraformaldehyde for 20 min and stained with 4′,6-diamidino-2phenylindole for 5 min. The cells were then sealed using the ProLong™ Gold Antifade reagent. Finally, the cells were imaged using confocal laser scanning microscopy (LSM 710 NLO; Carl Zeiss Meditec, Dublin, CA, USA) 31 .
For quantitative analysis, KU-19-19 cells at a density of 1 × 10 6 per well were seeded in 10 cm cell culture dishes and cultured for 24 h. The β-E, PEG-Lipo-β-E, and ATF 24 -PEG-Lipo-β-E were added to the culture media at a β-E concentration of 50 µg/mL, and incubated at 37 °C for 4 h. Then, the cells were trypsinized with a 0.25% trypsin-PBS solution and washed twice with ice-cold PBS, followed by a 3-cycle freeze-thaw procedure. Next, 600 µL of acetonitrile were added to 200 µL of the cell suspension with mixing for 10 min to precipitate the protein. After the centrifugation (9,400 × g for 10 min), the supernatant was collected, and the amount of β-E internalized by the KU-19-19 cells was detected using UFLC 32 .

In vitro cell migration
For the wound healing assay, the cells were seeded at a density of 2 × 10 6 cells per well in 6-well plates and cultured until 80%-90% confluency was reached. Then, a sterile 200 µL pipette tip was used to generate a vertical wound. The cells were washed with PBS 3 times and incubated with PEG-Lipo, PEG-Lipo-β-E, or ATF 24 -PEG-Lipo-β-E (at an equivalent β-E dose of 45 µg/mL). Cells that did not receive any treatment after the scratch were used as a negative control. In the ATF 24 + ATF 24 -PEG-Lipo-β-E group, the cells were incubated with excess free ATF 24 for 30 min after scratching. Then, ATF 24 -PEG-Lipo-β-E was added to the cells, and the following processes were performed as described for the other groups. Images were captured at 0 and 36 h after the treatment using an inverted microscope (Carl Zeiss, Jena, Germany) 33 .
For the Transwell migration assay, 2 × 10 5 cells were seeded in the upper chamber of Transwell plates (24-well insert, pore size: 8 µm; Corning, Corning, NY, USA). In total, 100 µL of serum-free media with PEG-Lipo, PEG-Lipo-β-E, or ATF 24 -PEG-Lipo-β-E (at an equivalent β-E dose of 45 µg/mL) was added to the upper chamber, while 600 µL of complete medium containing 10% FBS was loaded into the lower chamber as a chemoattractant. Cells without any treatment were used as controls. In the ATF 24 + ATF 24 -PEG-Lipo-β-E group, the cells were incubated with excess free ATF 24 for 30 min before adding the ATF 24 -PEG-Lipo-β-E. After 36 h, the upper medium was removed, and a cotton swab was used to remove the remaining cells. Then, the Transwell chambers were lightly washed with PBS, fixed using 4% paraformaldehyde for 20 min, and stained with Crystal Violet for 2 h. Finally, the upper chambers were washed with PBS twice, and the stained Transwell chambers were visualized using a microscope (Eclipse Ci-S; Nikon, Tokyo, Japan) 32 .

In vivo image studies
Female nude mice were injected with KU-19-19 cells to establish an orthotopic bladder cancer model. After 7 days of tumor establishment, the tumor-bearing mice were randomly assigned to 3 groups and received an intravenous injection of DiD, DiDlabeled PEGylated liposomes (DiD-PEG-Lipo), or DiD-labeled ATF 24 peptide-targeted liposomes (DiD-ATF 24 -PEG-Lipo). After anesthetization with 1% pentobarbital sodium, the mice were imaged with an IVIS® Lumina LT Series III in vivo imaging system (PerkinElmer, Waltham, MA, USA) at 24, 48, and 60 h after injection. Then, the mice were sacrificed, and the hearts, livers, spleens, lungs, kidneys, and tumors were collected. All organs were also imaged with the IVIS® Lumina LT Series III in vivo imaging system (PerkinElmer) 31 .

In vitro cytotoxicity studies
To determine the cell viability after incubation with PEG-Lipo-β-E or ATF 24 -PEG-Lipo-β-E and DDP, a CCK-8 assay was carried out using the KU-19-19 cells. The cells were seeded at a density of 3×10 4 cells per well in 96-well plates and incubated at 37 °C and 5% CO 2 for 24 h. Then, various concentrations of drugs were added to the wells. The control was RPMI 1640 medium without drugs. After 48 h of incubation, 10 µL of CCK-8 reagent was added to each well. After 2 h of incubation, the absorbance measurements were performed at 450 nm using a microplate reader (Multiskan™ FC; Thermo Fisher Scientific, Waltham, MA, USA) 34 .

Cell viability assay
The cells were treated with ATF 24 -PEG-Lipo-β-E (40 or 45 µg/mL) and/or DDP (1 or 2 µg/mL) for 48 h. The other steps were the same as described above. The combination index (CI) was measured by using CompuSyn software (ComboSyn, Paramus, USA).

Colony formation assay
The cells were first digested with 0.25% trypsin and then divided into individual cells. Subsequently, 5,000 cells were seeded into 10 cm culture dishes, grown overnight, and treated with ATF 24 -PEG-Lipo-β-E (8 µg/mL) and/or DDP (0.2 µg/mL) for 12 days. Once the colonies were visible to the naked eye, the culture dish was washed twice with PBS. Then, the colonies were fixed with 4% paraformaldehyde for 20 min, followed by staining with Crystal Violet for 2 h. Images were captured using a digital camera (EOS 5D Mark IV; Canon, Tokyo, Japan) 35 .

Cell apoptosis analysis
Cell apoptosis was assessed by flow cytometry using a CytoFLEX S (Beckman Coulter Biotechnology, Suzhou, China). Briefly, following the treatment with ATF 24 -PEG-Lipo-β-E (45 µg/mL) and/or DDP (1 µg/mL) for 48 h, the cells (3 × 10 5 cells/mL) were harvested and resuspended in 1× Annexin V-binding buffer. Subsequently, 4 µL of Annexin V-FITC and propidium iodide (PI) were added, and the cells were incubated for 10 min away from light at room temperature. Cell apoptosis was then analyzed using CytExpert software (Beckman Coulter, Brea, CA, USA) 35,36 .

Cell cycle analysis
Following treatment with ATF 24 -PEG-Lipo-β-E (45 µg/mL) and/or DDP (1 µg/mL) for 48 h, the cells (3 × 10 5 cells/mL) were collected, washed twice with ice-cold PBS, suspended in 300 µL of Reagent A DNA Staining Solution and 3 µL of Reagent B Permeabilization Solution, and incubated for 30 min at 37 °C in the dark. Cell cycle analysis was performed using CytExpert software (Beckman Coulter) 37 .

In vivo antitumor efficacy
The antitumor efficacy of ATF 24 -PEG-Lipo-β-E and the combined treatment of ATF 24 -PEG-Lipo-β-E and DDP injection were evaluated in tumor-bearing xenografts. In total, 25 female nude mice were used in this experiment, and each mouse was subcutaneously injected with a suspension of 5 × 10 5 KU-19-19 human bladder cancer cells in PBS (100 µL). When the tumor volume (TV) reached ~100 mm 3 , the mice were randomly divided into 5 groups (N = 5) and subjected to one of the following treatments: (a) PEG-Lipo, (b) PEG-Lipo-β-E (25 mg/kg), (c) ATF 24 -PEG-Lipo-β-E (25 mg/kg) via intravenous injection daily, (d) DDP injection (5 mg/kg), or (e) ATF 24 -PEG-Lipo-β-E (25 mg/kg) + DDP injection (5 mg/kg) intravenously twice 39 . The body weights were recorded, and tumor growth was monitored every 4 days. Sixteen days later, all animals were sacrificed by cervical dislocation, and the tumors were isolated and weighed. The TV was calculated by the following formula: TV = 0.5× (d1 × d2 2 ), where d1 and d2 are the largest and the smallest perpendicular diameters, respectively. After the livers, spleens, and kidneys were harvested, hematoxylin and eosin (H&E) staining was used to evaluate the safety of the various formulations. Then, the tumor tissues were evaluated qualitatively to determine uPAR, Ki-67, and cleaved caspase-3 expressions using immunohistochemistry 40 .

Statistical analysis
The data were statistically analyzed by Prism 7 software (GraphPad, San Diego, CA, USA). All analyses were conducted in triplicate, and the results are reported as the mean ± standard deviation (SD). A value of P < 0.05 was considered statistically significant (*P < 0.05; **P < 0.01; ***P < 0.001).

Synthesis of DSPE-PEG 2000 -ATF 24
The ATF 24 peptide Cys-Leu-Asn-Gly-Gly-Thr-Cys-Val-Ser-Asn-Lys-Tyr-Phe-Ser-Asn-Ile-His-Trp-Cys-Asn-Cys-Pro-Lys-Lys (purity > 97.16%) was synthesized according to standard solid phase peptide synthesis by ChinaPeptides (Shanghai, China) 19 . The theoretical MWs of ATF 24 and DSPE-PEG 2000 -COOH were 2,717 and 2,750 Da, respectively 41 . As shown in Figure 2B and 2C, the MW of the ATF 24 peptide as determined by LC-ESI-MS was 2,717 Da, and the MW of the reaction product as determined by MALDI-TOF-MS was approximately 5,143 Da, which was consistent with the theoretical MW of ATF 24 and DSPE-PEG 2000 -ATF 24 , confirming that the synthetic product was the targeted compound.

Preparation and characterization of ATF 24 -PEG-Lipo-β-E
A schematic diagram of ATF 24 -PEG-Lipo-β-E is shown in Figure 2A. images generally displayed a spheroid shape with a size of 79.32 ± 1.282 nm (Figure 2D and 2E), and the PDI value was 0.28 ± 0.008, indicating a homogenous population of phospholipid vesicles. The zeta potential was -12.77 ± 0.416 mV, indicating a negatively charged surface on the liposomes (Figure 2F). The pH was 6.49 ± 0.017. The β-E content was 5.24 ± 0.362 mg/ mL. The average EE of β-E in the ATF 24 -modified liposomes was 98.37 ± 0.95%.

FTIR and DSC
The FTIR spectra obtained for the pure drug, physical mixture, and liposomes are shown in Figure 2G. In the spectra, the bands at 1739 and 1639 cm −1 indicated the presence of the C=C, C=O, and NH 2 in the β-E, DSPE, soybean lecithin, and ATF 24 .
The asymmetric stretching vibration of the P=O group was identified at 1238 cm −1 and 1231 cm −1 in the physical mixture and liposomes. In addition, the peaks at 2856 and 2927 cm −1 were related to the symmetric and asymmetric CH 2 stretching vibrations in the aliphatic structure. The two peaks related to the β-E (3082 -1 and 889 cm -1 ) were not observed in the spectra of PEG-Lipo-β-E and ATF 24 -PEG-Lipo-β-E likely because the drug was encapsulated in the liposomes. Moreover, the spectra of ATF 24 -PEG-Lipo-β-E were similar to those of PEG-Lipo and PEG-Lipo-β-E, indicating that the presence of the ATF 24 peptide did not affect the formation of the liposomes 42 .
A thermogram of ATF 24 -PEG-Lipo-β-E is shown in Figure 2H. Our   peak at temperatures of 36.83 °C, 42.34 °C, and 45.11 °C, respectively. The DSC curve of ATF 24 -PEG-Lipo-β-E exhibited a high intensity sharp endothermic peak at 46.27 °C, which significantly differed from the peaks of the soybean lecithin, cholesterol, and the physical mixture. This peak was also detected in the thermograms of PEG-Lipo and PEG-Lipo-β-E at the temperatures of 45.99 °C and 46.14 °C, although the peak became less intense. These changes were attributed to a possible interaction between the liposomal components 42 .

In vitro release of β-E from ATF 24 -modified liposomes
As shown in Figure 2I, PEG-Lipo-β-E and ATF 24 -PEG-Lipoβ-E exhibited a very slow release of β-E without an initial burst in 75% ethanol. Approximately 37% of the drug was released within the first 2 h; then, a sustained drug release profile was achieved in which 47% of the drug was released after 8 h, while free β-E displayed approximately 8.94% release within 8 h under these conditions. The release of β-E significantly increased after being encapsulated into the liposomes, but did not increase after the addition of ATF 24 , indicating that modification of the ATF 24 peptide did not affect the release characteristics of the PEGylated liposomes.

Receptor expression
The expression level of uPAR was detected using western blot analysis. As shown in Figure 3A, the highly invasive bladder cancer cell line, KU-19-19, had a higher expression level of uPAR than the less aggressive cell line, RT4.

Cellular uptake study
To assess the cellular uptake and cancer-targeting ability of the ATF 24 -modified liposomes in vitro, KU-19-19 cells were cultured with DiD-PEG-Lipo or DiD-ATF 24 -PEG-Lipo. The ATF 24 -modified formulation showed a higher uptake than the PEGylated formulation, as shown in the fluorescent images of the KU-19-19 cell lines after incubation for 4 h (Figure 3B). The cellular uptakes of the free β-E, PEG-Lipo-β-E, and ATF 24 -PEG-Lipo-β-E were quantitatively evaluated by using UFLC. ATF 24 -PEG-Lipo-β-E displayed the highest cellular uptake. The uptake of free β-E was significantly lower than that of PEG-Lipo-β-E. Compared to the PEG-Lipo-β-E group, ATF 24 -PEG-Lipo-β-E increased the uptake amount of β-E by almost 1.34-fold ( Figure 3C).

In vitro cell migration
Because cell migration is closely related to tumor metastasis, the cell motility was examined using a wound healing assay.
In the control group, the scratch gap was barely observed after 36 h, indicating that the KU-19-19 cells had strong motility. Blank liposomes without β-E did not affect cell motility. The PEG-Lipo-β-E groups prevented cell motility with 54.33% wound closure. The wound closure in the ATF 24 -PEG-Lipoβ-E group was reduced to only 25.67%, showing the highest inhibition of cell motility. After preincubation with ATF 24 , wound closure could not be efficiently suppressed by ATF 24 -PEG-Lipo-β-E (Figure 3D and 3E).
In the Transwell migration assay, PEG-Lipo had almost no inhibitory effect, but the inhibitory effect of PEG-Lipo-β-E was significant. Compared with the PEG-Lipo-β-E group, the migrated cells were further reduced in the ATF 24 -PEG-Lipo-β-E group. Moreover, after incubation with excess free ATF 24 peptide, the migration of KU-19-19 cells could not be efficiently suppressed by ATF 24 -PEG-Lipo-β-E, which explained the specific binding of ATF 24 and uPAR (Figure 3D  and 3F). As shown in this figure, the liposomes decorated by ATF 24 effectively inhibited tumor cell migration.

In vivo image studies
The KU-19-19 orthotopic bladder cancer model was successfully developed in nude mice. Figure 4A shows in vivo images of the KU-19-19 bladder cancer-bearing mice after an intravenous injection of free DiD or DiD-labeled liposomes. The free DiD group only showed fluorescence intensity in the liver and did not show accumulation in the tumor site, while the DiDlabeled liposomes showed significant fluorescence intensity in the liver and tumor site. The DiD signals of DiD-PEG-Lipo observed at 24, 48, and 60 h were weaker than those of DiD-ATF 24 -PEG-Lipo. In the DiD-ATF 24 -PEG-Lipo group, the fluorescence intensity at 48 h after injection was the strongest among the three time points. The ex vivo organs and tumors are shown in Figure 4B and 4C; free DiD, DiD-PEG-Lip, and DiD-ATF 24 -PEG-Lipo mainly accumulated in the liver and spleen, and the DiD fluorescence in the tumors in the DiD-ATF 24 -PEG-Lipo group was stronger than that in the DiD-PEG-Lipo group.

In vitro cytotoxicity studies
As shown in Figure 5A and   the enhanced killing effect was not obvious when compared with that of PEG-Lipo-β-E, which might be due to the small number of cells.

ATF 24 -PEG-Lipo-β-E and DDP synergistically inhibit the viability and proliferation of bladder cancer cells
β-E has been shown to enhance DDP cytotoxicity in carcinomas of the bladder 43 . In this study, we investigated whether the combination of ATF 24 -PEG-Lipo-β-E and DDP had a synergistic effect. The viability percentages of the combination groups were significantly lower than that in each single drug group, indicating the greater toxicity efficacy of the combination therapy ( Figure 5C). To further determine whether the improved cytotoxicity was the result of synergism or simply an additive effect by two drugs, the CI was calculated to quantitatively define synergism (CI < 0.90), an additive effect (CI = 0.90-1.10), or antagonism (CI > 1.10) in drug combinations 44,45 . The CI values of the three combination groups were below 0.90, indicating a synergistic effect of ATF 24 -PEG-Lipoβ-E and DDP ( Table 1). The flow cytometry analysis indicated that DDP, in combination with ATF 24 -PEG-Lipo-β-E, caused an elevation in cellular apoptosis (Figure 5D, 5E, and 5H). We then assessed the effect of the combinatory treatment on cellular proliferation, so the cell cycles of cancer cells were first analyzed. The results showed that, compared with either single agent, the combination of ATF 24 -PEG-Lipo-β-E and DDP caused significant cell cycle arrest at the G2/M phase ( Figure  5F and 5I). Consistent with this finding, the colony-forming ability of the KU-19-19 cells was significantly reduced by co-treatment of ATF 24 -PEG-Lipo-β-E and DDP, compared with those observed following treatment with either single agent treatment (Figure 5G) 46 . Together, the results indicated that the co-treatment of ATF 24 -PEG-Lipo-β-E and DDP had a synergistic effect in reducing cellular viability and proliferation in bladder cancer.

ATF 24 -PEG-Lipo-β-E increased DDP cell apoptosis and cell cycle arrest by the caspasedependent pathway and the Cdc25C/Cdc2/ cyclin B1 signaling pathway
Western blot was used to detect the levels of key proteins in the caspase-dependent pathway, and key proteins at the G2/M phase, in KU-19-19 cells treated with DDP combined with ATF 24 -PEG-Lipo-β-E for 48 h. The results showed that compared to the DDP treatment group, DDP combined with ATF 24 -PEG-Lipo-β-E decreased the expressions of Bcl-2, Cdc25C, Cdc2, and cyclin B1, but significantly increased the expressions of cleaved PARP, cleaved caspase-3, and Bax, when compared to the DDP treatment group (Figure 6) 47,48 .

Combined treatment with ATF 24 -PEG-Lipo-β-E and DDP injection enhanced the antitumor effects in a bladder cancer xenograft model
As shown in Figure 7A and 7B, PEG-Lipo-β-E exhibited a high antitumor efficacy with an average TV of 784.77 ± 82.61 mm 3 . The antitumor efficacy of PEG-Lipo-β-E was augmented with an average TV of 544.95 ± 54.75 mm 3 after conjugating the ATF 24 peptide onto the liposomes, facilitated by its active targeting performance. Moreover, ATF 24 -PEG-Lipo-β-E further augmented the antitumor efficacy of the DDP injection through its sensitizing effect. The combination therapy resulted in profound tumor regression (74.18% decrease, as detected by TV), while the DDP injection treatment caused an approximate 58.75% decrease in the TV. Analyses of the tumor weights also confirmed that compared with the single agent treatments, the tumor weight was reduced by the drug combination ( Figure 7C). As seen in Figure 7D, there was no significant decrease in body weight of mice during the treatment.
Based on the H&E staining results (Figure 8A), we found that the hepatocytes and splenocytes were normal after drug treatment, and that the kidney exhibited the normal architecture of glomeruli and renal tubules without vascular expansion or congestion. No hepatorenal toxicity was observed by H&E staining in any group. Based on the immunochemical staining shown in Figure 8B, it was concluded that the ATF 24 -PEG-Lipo-β-E group did not show positive expression of uPAR, while the control group and PEG-Lipoβ-E group showed positive areas of uPAR. We also observed that the expression of uPAR in the combined group was lower than that in the DDP group. Proliferation is a key feature of the progression of tumors and is currently widely estimated to occur because of the Ki-67 nuclear antigen 49 . The expression of Ki-67 in the combined group was lower than that in the other three groups, and the control group had large positive areas of Ki-67. Caspase-3 is a type of protein that can be activated in apoptotic cells. We observed that the expression of cleaved caspase-3 in the combined group was higher than that in the other three groups. The control group had dense nuclei accumulation, but few cleaved caspase-3-positive spots. Overall, treatment with ATF 24 -PEG-Lipo-β-E and DDP injection prevented the KU-19-19 bladder cancer cells from further proliferation, and simultaneously induced apoptosis in cancer cells.

Discussion
Over the past 30 years, drug delivery systems using liposomes have attracted attention due to their potential to improve the pharmacokinetics and biodistribution of free drugs and function as drug reservoirs. With small particle sizes, liposomes can passively target tumor tissues via the EPR effect. Liposomes can also be used to actively target tumors using ligands modified on liposomal surfaces [50][51][52] . However, the low delivery efficiency and limited tumor penetration of nanoparticle-based drug delivery systems are still considered an "Achilles heel" in tumor treatment 53 . Therefore, it is likely that tumor-targeting strategies enabling the nanoparticle carriers to simultaneously target the tumor microenvironment and tumor cells offer promising targeted drug delivery approaches. Highly invasive bladder cancer does not have effective anticancer targets for clinical treatments. However, the overexpression of uPAR proteins has been observed in MIBC and several stromal cell types in the tumor microenvironment, providing a potential strategy for targeting MIBC 54 . Studies have indicated that the co-delivery of chemotherapeutic drugs using uPAR-targeted nanoparticles enhanced the therapeutic efficacy in triple negative breast cancer and pancreatic cancer animal tumor models. Currently, in clinical studies, β-E has been shown to be a promising adjunctive treatment, exerting a synergistic effect and improving patient outcomes in the treatment of cancer 55 . In this study, we developed uPAR-targeted liposomes carrying β-E to overcome the physical barrier of the stroma for the effective treatment of MIBC.
The uPAR-targeted liposome ingredient, DSPE-PEG 2000 -ATF 24 , was successfully synthesized by conjugating ATF 24 , the ATF peptide of the receptor-binding domain of uPA, a natural ligand of uPAR, with DSPE-PEG 2000 -COOH. ATF 24 -modified liposomes were successfully prepared with a high EE and a uniform size. The content of β-E in the ATF 24 modified liposomes was equal to that of the commercial elemene injection. The FTIR and DSC analyses also demonstrated the formation of liposomes. Notably, the liposomes prepared in this study were characterized by TEM, which could not reflect the true appearance of the sample, and which is a shortcoming of this study, so the ATF 24 -PEG-Lipo-β-E needs to be characterized by cryo-TEM in subsequent studies. The cellular uptake efficacy of liposomes was studied in the KU-19-19 cell line by confocal microscopy and UFLC. The fluorescence intensity after incubation with DiD-ATF 24 -PEG-Lipo was greater than that after incubation with DiD-PEG-Lipo. The cellular uptake as determined by UFLC was consistent with that in the fluorescence intensity analysis. Given the important role of uPAR in regulating matrix degradation, metastasis, and angiogenesis, we also compared the ability of liposomes to inhibit cell migration, by using wound healing and Transwell migration assays. KU-19-19 cells incubated with ATF 24 -PEG-Lipo-β-E produced the most effective results. Preincubation with excess free ATF 24 peptide lowered the inhibitory effect of ATF 24 -PEG-Lipo-β-E. Notably, although the above results initially indicated that ATF 24 -PEG-Lipo-β-E had good targeting, the experimental design was not rigorous, and we did not use a scramble peptide sequence as control materials, which is a shortcoming in this study that needs to be addressed in subsequent research.
The biodistribution of ATF 24 -PEG-Lipo was also investigated in the KU-19-19 orthotopic bladder cancer model. Notably, fluorescence in the tumor tissues in the DiD-PEG-Lipo group gradually reduced from 24 to 60 h, while the fluorescence in the DiD-ATF 24 -PEG-Lipo group gradually increased up to 48 h, and then decreased until the end of the experiment. Additionally, the DiD signals in the liver and spleen were greater than those in the other organs, which might be due to capture by the reticuloendothelial system. The low accumulation of both liposomes in the heart also indicated no obvious heart toxicity using these liposomal delivery systems. Taken together, these results showed that ATF 24 modification increased the tumor-specific delivery of ATF 24 -PEG-Lipo in vivo, likely resulting from the reduced drug flow back to circulation via an interaction between ATF 24 in the liposomes and the uPAR expressed on the cell surface 56 .
DDP, which is among the most potent chemotherapeutic drugs, has activity against a variety of solid tumors 55 . Gemcitabine plus DDP is the neoadjuvant regimen of choice in many institutions treating patients with MIBC, but can be limited by severe side effects and drug resistance to the chemotherapeutic agents 57 . β-E enhanced the efficacy and reduced the toxicity of chemoradiotherapy and reversed the drug resistance. In the present study, our results showed that the combination of ATF 24 -PEG-Lipo-β-E and DDP significantly increased cell apoptosis, and arrested the cell cycle at the G2/M phase by the caspase-dependent pathway and the Cdc25C/Cdc2/cyclin B1 signaling pathway. Notably, the ATF 24 -PEG-Lipo-β-E showed a much higher level of apoptosis than DDP; however, the changes in the apoptosis-related proteins treated with ATF 24 -PEG-Lipoβ-E alone were not more obvious than those in the DDP treatment group. Apoptosis-related proteins (cleaved PARP, cleaved caspase-3, Bax, and Bcl-2) treated with ATF 24 -PEG-Lipo-β-E + DDP had a significant change compared with the DDP alone group. It might be possible that ATF 24 -PEG-Lipo-β-E did not primarily induce cell apoptosis via the mitochondrial pathway, but it could increase the sensitivity of tumor cells to DDP. The antitumor mechanism of ATF 24 -PEG-Lipo-β-E and the synergistic anti-tumor mechanism of the two drugs need to be further studied in the future.
The hepatorenal toxicity of DDP and the β-E liposomal formulations were important concerns in this study, because DDP and β-E are mainly metabolized by the liver and excreted by the kidneys, and the liposomes were found to be highly accumulated in the liver and spleen, as revealed by an in vivo imaging study. Therefore, the toxicity of the various formulations was assessed by histological tissue imaging of three organs (the liver, spleen, and kidneys) using H&E staining. There were no evident pathological abnormality in the three organs after the treatment with ATF 24 -PEG-Lipo-β-E and DDP injection. This finding confirmed the lack of toxicity to the major organs after combined treatment.
In the pathological analysis of tumor tissues, we detected three types of related cytokines, i.e., uPAR, Ki-67, and cleaved caspase-3. The low expression of Ki-67 indicated reduced growth of tumor cells. In addition, the high expression of cleaved caspase-3 in tumor sites indicated the occurrence of tumor cell apoptosis. Based on pathological analysis, we concluded that ATF 24 -PEG-Lipo-β-E and DDP injection effectively inhibited the proliferation of cancer cells, and induced cancer cells to undergo apoptosis. These results may be related to the decreased expression of uPAR.

Conclusions
Overall, we successfully developed a novel ATF 24 peptidetargeted liposome carrying β-E for targeted delivery into uPAR-overexpressing bladder cancer cells. The present study also provided an effective combined strategy for bladder cancer treatment.