The Single-cell sequencing (scRNA-seq) data of BC (GSE135337) was download from the Gene Expression Omnibus database (GEO) (https://www.ncbi.nlm.nih.gov/geo/). TCGA-BLCA dataset download from The Cancer Genome Atlas Program (TCGA) (portal.gdc.cancer.gov). The flow chart of this study is shown in Figure 1 .

ScRNA-seq were de-batched using the R package “harmony” (1.2.0) function “harmony”, space was further projected into a 2-dimensional space for visualization by Uniform Manifold Approximation and Projection (UMAP, R package: Seurat 4.4.4), then using resolution=0.5 for cell clustering. and labelled with broadly marker epithelial (EPCAM, KRT13, KRT7, KRT18), fibroblast (DCN, COL3A1, COL1A1), immune (PTPRC), or endothelial (PLVAP, VWF, CLDN5) using a panel of marker genes gleaned from the literature.

DEGs were identified using the function “FindMarkers” with a threshold of |log2(FC)| > 0.25 and an adjusted P value < 0.05.

Kyoto Encyclopedia of Genes and Genome (KEGG), Gene Ontology (GO) enrichment analysis and Gene Set Enrichment Analysis (GSEA) were performed for differential genes through the R package “clusterProfiler” (4.10.0), and the enrichment results with P value < 0.05 were selected.

A total of 122 BC tissues and 34 non-tumor bladder tissues were collected from BC patients at The Second Affiliated Hospital of Nanchang University between 2019 and 2023. All patients did not receive radiotherapy, chemotherapy, or immunotherapy prior to surgical treatment and were pathologically diagnosed with BC. All methods were carried out in accordance with relevant guidelines and regulations, and all experimental protocols were approved by the Ethics Committee of the Second Affiliated Hospital of Nanchang University (Approval No.: IIT- 2024-228). Confirming that informed consent was obtained from all subjects or their legal guardian.

Collected tissues were fixed with 4% paraformaldehyde for 4 hours, sliced into 3.5 μm after paraffin embedding. Then after deparaffinization and rehydration, heat-induced antigen retrieval with sodium citrate was performed for 20 min. Followed by quenching of endogenous peroxidase activity though 3% hydrogen peroxide for 5 min. Primary antibody incubation with DUSP6 (Santa Cruz Biotechnology, sc-377070, 1:50), KI-67(Abmart Inc, TW0001, 1:50), PINK1(Santa Cruz Biotechnology, sc-517353, 1:50), PARKIN (WANLEIBIO, WL02512, 1:100) was performed overnight at 4°C. Flowed by hematoxylin, biotinylated secondary antibody, and 3,3’-diaminobenzidine (DAB) solution (Zhongshan Golden Bridge Biotechnology Co Ltd, ShangDong, China) were then used for standing.

The human BC cell line T24 was obtained from the China Academia Sinica Cell Repository (Shanghai, China), and the UC3 cell line was acquired from Procell (Wuhan, China) with STR profiling. T24 cells were cultured in DMEM (Servicebio, Wuhan, China), and UC3 cells were cultured in MEM (Servicebio, Wuhan, China), both supplemented with 10% fetal bovine serum (Excell, Suzhou, China). Plasmids LV-shNC, LV-shDUSP6-1, LV-shDUSP6-2, and LV-shDUSP6-3 (SH1: GGAGAACGCAGGAGAGTTTAAA; SH2: GCTGTGGTGTCTTGGTACATTG; SH3: AAACTGTGGTGTCTTGGTACAT) purchase from Qingke (Beijing Qingke Biotechnology Co., Ltd.) were used to construct stable cell line.

A total of 20 μg plasmids PxpAx2, PMD2g, and PLVX were mixed at ratio of 3:1:4 in 500 μl of serum-antibiotic-free DMEM. 50 μg of PEI (Yeasen Biotechnology (Shanghai) Co., Ltd.) was diluted in 500 μl of the same medium. The two solutions were mixed, incubated at room temperature for 20 minutes, and added to 293T cells in a 10 cm dish. After 18 hours, the medium was replaced with complete medium, and the cell supernatant was collected at 48 and 72 hours for subsequent cell infection.

Then bladder cancer cells in logarithmic growth phase were transfected by lentivirus, following by culturing in 1640/MEM/DMEM medium with 10% FBS in a 6-well dish. Puromycin (2 μg/μl) was added for selection when the cell density reached 90%, and the stable colonies will be amplified after 10–14 days. The overexpression/knockdown efficiency of TEAD4 was evaluated by qPCR and western blot.

The cell was lysed in RIPA buffer (Epizyme Biomedical Technology, shanghai, China), and electrophoresis was using 10% SDS-PAGE (Epizyme Biomedical Technology, shanghai, China) then transferred protein into PVDF membranes (Merck Millipore Burlington, USA), after 1.5 h block incubate with first antibodies: DUSP6 (Santa Cruz Biotechnology, sc-377070, 1:200), PINK1 (Santa Cruz Biotechnology, sc-517353 1:500),P62 (Abmart Inc, T55546, 1:5000),LC3 (Abmart Inc, T55992, 1:500), mTOR (Abmart Inc, T56571 1:1000), P-mTOR (Abmart Inc, T56571 1:1000), PARKIN (WANLEIBIO, WL02512, 1:1000), ACTIN (Servicebio, GB113225-100, 1:1000) for a night at 4°C. Next day, incubated with second antibody (Servicebio, GB23301 1:10000, GB23303 1:10000) at room temperature for 1.5 h. Finally, the blots were incubated with ECL reagents (Servicebio, Wuhan, China) and visualized using Image J software.

Real-time quantitative PCR assay using Universal SYBR Green qPCR Master Mix (Servicebio, Wuhan, China). The primer purchased from Sangon (Sangon Biotech (Shanghai) Co., Ltd.) and sequences are as follows:DUSP6: Forword:5′- GAACTGTGGTGTCTTGGTACATT -3′; Reverse: 5′- GTTCATCGACAGATTGAGCTTCT -3′; β-ACTIN: Forword: 5′ - CATGTACGTTGCTATCCAGGC -3 ′; Reverse: 5′- CTCCTTAATGTCACGCACGAT -3 ′. The program as follows: 1. Initial denaturation for 3 min. 2. Denaturation for 20 second. 3. Annealing and extension for 20 seconds.

500 cells were seeded in 6-well plates with culture medium containing 10% FBS and cultured for 1–2 weeks. The cell colonies were fixed with 4% paraformaldehyde for 30 minutes, then staining with 0.1% crystal violet for an additional 30 minutes. Finally, images were captured using a high-definition digital camera and analyzed using ImageJ software.

Cells were seeded into 12-well plates at a density of 10,000 cells per well and maintained at 37°C in an atmosphere of 5% CO2. At 2, 4, and 6 days, the cells were digested, resuspended in total 500 µL of medium, and the cell density was measured using a flow cytometer.

Transwell chambers (Labselect, 8 μm pore size, Anhui, China) with or without Matrigel was used to evaluate the invasion or migration ability of BLCA cells. The procedure involved placing 2 × 10^5 cells into the upper compartment, which was filled with 200 μL of a serum-free medium. In the lower compartment, 500 μL of medium supplemented with 10% FBS was introduced. Following an incubation period of 16 to 24 hours at 37°C, the Transwell chambers were rinsed with PBS and treated with 4% paraformaldehyde for about 15 minutes. Subsequently, the cells on the upper surface of the membrane were wiped off using a cotton swab, and the remaining cells were stained with crystal violet for approximately 15 minutes at ambient temperature. Finally, the membranes were rinsed again in PBS, allowed to air-dry, and then documented with photographs.

A 6-well plate was used for cell seeding when conducting the wound-healing assay. A 200μL sterile plastic pipette was employed to create a scratch in the cell monolayer. Following this, the cells were placed in a medium devoid of FBS for cultivation. Photographs were taken at two time points, 0 and 24 hours, utilizing an electron microscope. The cells’ migratory capacity was assessed through the quantification of the wound area’s dimensional alterations.

Four-week-old male nude mice were purchased from SPF (SPF biotechnology, Beijing) biotechnology. Transfected cells were subcutaneously injected (1 × 10^6 in 200 μL Matrigel) in 5 mice per group. Four weeks later, after anesthesia with sodium phenobarbital, the mice were euthanized, and the tumor tissues were harvested for subsequent experiments. And all methods were carried out in accordance with relevant guidelines and regulations, and all experimental protocols were approved by Institutional Animal Care and Use Committee, Nanchang University (Approval No.: NCULAE-20221031184).

Cells or tissue were fixed in 2.5% glutaraldehyde (Servicebio, Wuhan, China) at 4 °C for 4 h, pre-embedded in 1% agar to maintain integrity, washed with PBS, and thereafter post-fixed for 2 h in the presence of a 1% OsO4 buffer at 4°C. After washing, cells were dehydrated in graded ethanol concentrations and embedded using Epon812 epoxy resin. Ultrathin sections (90 nm) were obtained on copper grids, double-stained with 0.2% lead citrate and 1% uranyl acetate, followed by examination by HT7800 transmission electron microscopy (Hitachi, Japan).

ROS: Diluted DCFH-DA (Beyotime, Shanghai) with serum-free culture medium (1:1000). After trypsinizing and counting the cells, take out 500,000 cells, centrifuge and remove the medium, resuspend the cells with diluted DCFH-DA. Incubate at 37°C cell culture incubator for 20 minutes. Then using flow cytometer with an excitation wavelength (Ex) of 488nm and an emission wavelength (Em) of 525nm to detect the fluorescence.

ATP: Dilute the ATP detection reagent (Beyotime, Shanghai) with diluent at ratio of 1:4. and add 200 μL of lysis solution to 2 million cells. After lysing, centrifuge at 12,000g for 5 minutes at 4°C, the supernatant was used for subsequent assay. Add 100 microliters of ATP working solution to 96-well plate, stand at room temperature for 3–5 minutes before adding 20 microliters of lysate supernatant. At last, measure the relative light units (RLU) though multifunctional microplate reader.

Mitochondrial membrane potential: Dilute JC-1(Beyotime, Shanghai) with staining buffer at ratio of 1:200, add 500 microliters culture medium and 500 microliters diluted working solution to resuspend the cells. then Incubate it at 37°C for 20 minutes and centrifuge at 600g for 3 minutes at 4°C wash the cells three times with PBS before resuspending in buffer. Finally, measure the fluorescence intensity of JC-1 monomers (Ex:485 Em:535) and JC-1 aggregates (Ex:550 Em:600) though flow cytometer.

Autophagy: Trypsinizing and seeding cells to ensure that the cell density is approximately 50% in the next day. Discard the old culture medium and add fresh medium The mCherry-GFP-LC3B (Genechem, Shanghai) adenovirus added according to MOI = 10. After a two-day incubation, the cells were imaged by confocal microscope to visualize autophagy flux.

Before being collected for subsequent experiments, cells were pretreated with complete medium containing 0.5 μM mitophagy inhibiter ciclosporin A (CSA, MedChemExpress) for 24 hours.

Cells were treated with a complete medium containing 1.5 μM mTOR agonist MHY1485 (MedChemExpress) for 12 hours, after that they were collected for subsequent experiments.

Unless otherwise stated, assays were repeated thrice, and all data showed as independent data points. SPSS 25.0 was used for data analyses. For quantitative data, normality testing is used to ascertain whether the data is normally distributed, and then a two-tailed Student’s t-test or non-parametric test is utilized. The χ2 test or Fisher’s exact test is conducted for qualitative data. P < 0.05 is accepted as being statistically significant.

In our previous study, while RAB14 was knocked down, we found the expression of DUSP6 was significantly elevated in BC cells (18). In order to explore the expression of DUSP6 in BC tissues, 122 tumor tissues and 34 non-tumor tissues were used for immunohistochemical (IHC) analysis. The findings indicated that the immune scores of tumor tissues were higher than those of non-tumor tissues ( Figure 2A ). Additionally, the staining intensity of MIBC was observed to be stronger than NMIBC ( Figures 2A, B ). Then, the results of the clinical correlation analysis indicated that the expression level of DUSP6 was positively correlated with T stage, N stage, and grade ( Table 1 ). However, no correlation was observed between DUSP6 expression and age, sex, M stage, or tumor differentiation ( Table 1 ).

A five-year follow-up survey of 122 BC patients revealed that the median survival time for patients with low DUSP6 expression was 46.8 weeks, which was significantly longer than the 37.7 weeks observed for patients with high DUSP6 expression. Kaplan-Meier survival analysis highlighting the potential of DUSP6 as a prognostic marker for patient survival in BC (p < 0.05). ( Figure 2C ).

The basic expression level of DUSP6 in BC cell lines was examined by qRT-PCR and Western blot. The results demonstrated that the mRNA and protein expression levels of DUSP6 in UC3, 5637, and T24 were significantly higher than those in the normal cell line SV-HUC-1( Figure 3A ). In accordance with the aforementioned results, UC3 and T24 cells exhibiting elevated endogenous DUSP6 expression were selected for the in vitro knockdown assay. The knockdown efficiency was determined by qRT-PCR, which revealed that shRNA#1 and shRNA#2 exhibited superior knockdown efficiency compared to shRNA#3 (p<0.001, Figure 3B ). Then, we investigated the proliferative, invasive, and metastatic capabilities of stable cell lines with DUSP6 knockdown (shRNA#1&shRNA#2). The results of Wound healing and Transwell assays found DUSP6 knockdown markedly hindered the invasive and migratory activities of BC cells ( Figures 3C, D ). Additionally, silencing DUSP6 also suppressed proliferation ( Figure 3E ) and colony formation ( Figure 3F ).

In the in vivo experiments utilizing a xenograft tumor model of nude mouse with subcutaneously injected T24 cells ( Figure 4A ), both the tumor volume and weight were decreased in the DUSP6 knockdown group compared to the control group ( Figures 4B–D ). Immunohistochemical examination of Ki67 staining demonstrated that DUSP6 knockdown inhibited tumor cell growth in vivo ( Figures 4E, F ).

In order to investigate the mechanism by which DUSP6 influences the malignant progression of BC. We preprocessed and annotated single-cell seq data and found DUSP6 show a high expression in endothelial cells and certain epithelial clusters ( Figure 5A , Supplementary Figures 1A-D ). Then the epithelial cells were extracted ( Figure 5B ). Cluster 3, 6, 7, 9, 10, 12, 13, 14 displayed higher expression levels of DUSP6 and were defined as DUSP6+ epithelial cells ( Figures 5C, D ). Subsequently, a differential expression analysis was conducted on DUSP6+ epithelial cells in comparison to other epithelial cells, resulting in the identification of 534 differentially expressed genes (DEGs) ( Figure 5E , Supplementary Table 1 ). An KEGG pathway enrichment analysis of these DEGs revealed that the enrichment pathways were p53 signaling pathway, mitophagy pathway, ferroptosis and apoptosis ( Figure 5F ). The biological process analysis further showed DUSP6 may influence the mitochondrial function ( Figure 5G ). In the independent external validation set TCGA, we also found that DUSP6 is associated with mitophagy and multiple mitochondrial functions ( Supplementary Figures 2A-C , Supplementary Table 2 ).

These findings demonstrate that DUSP6 is associated with multiple tumor-related processes, including the p53 signaling pathway, mitophagy, ferroptosis, and apoptosis. Among these, mitophagy was selected for further analysis due to its particular relevance to DUSP6.

In order to verify the relationship between DUSP6 and autophagy, we first analyzed the oxygen species (ROS), adenosine triphosphate (ATP), and mitochondrial membrane potential levels in DUSP6 knockdown cells. The results showed that shRNA-DUSP6 knockdown in BC cells result in significantly increase of intracellular ROS and decrease in ATP levels ( Figures 6A, B ). We also found a significant decrease in mitochondrial membrane potential in DUSP6 knockdown stable cells ( Figure 6C ). These results indicated that mitochondrial dysfunction occurred after DUSP6 knockdown. Additionally, enhanced autophagic flux and increased the number of autophagosomes within the cells were also found in DUSP6 knockdown cells ( Figures 6D, E ). This was accompanied by changes in the expression of mitophagy markers. The result of western blot indicated silencing DUSP6 could down-regulate P62 and up-regulate PARKIN, PINK1 and LC3II/I ( Figure 6F ). Then, we performed IHC staining to analyze PINK1 and PARKIN expression in xenograft tumor sample from nude mice. Similar to our previous results, it also demonstrated a significant elevation in PINK1 and PARKIN expression in tumor tissues subjected to DUSP6 knockdown ( Figure 6G ). The results of transmission electron microscopy provided further confirmation of the presence of elevated mitophagy ( Figure 6H ).

To elucidate whether mitophagy plays a role in the regulation of the malignant phenotype of invasion, metastasis, and proliferation in BC by DUSP6, we pre-treated cells with the mitophagy inhibitor CsA with 0.5 μM for 24 hours and performed functional phenotype assays. the results showed the CsA reversed the effects of DUSP6 knockdown on cell migration, invasion ( Figures 6I, K ), and proliferation partially ( Figure 6J ).

The findings indicate that DUSP6 may contribute to the malignant progression of BC by influencing mitophagy.

To investigate the potential mechanisms through which DUSP6 influences mitochondrial autophagy, we conducted Gene Set Enrichment Analysis (GSEA). The results indicated the existence of a notable positive correlation between DUSP6 and the mTOR pathway. ( Figures 7A, B ). Several studies have demonstrated that this pathway is associated with mitochondrial biogenesis and mitochondrial function (19–23). To substantiate the hypothesis that DUSP6 regulates the mTOR pathway, western blot analysis was conducted to examine alterations in mTOR phosphorylation levels in DUSP6 knockdown stable cells. The results demonstrated a reduction in the levels of mTOR phosphorylation ( Figure 7C ), Similarly, in the xenograft tumor samples, the p-mTOR level was found to be reduced in the sh-DUSP6 group ( Figure 7D ).

To further confirm whether DUSP6 mediates mitophagy in bladder cancer through the mTOR pathway, we pre-treated DUSP6 knockdown cells with medium containing 1.5 μM of mTOR agonist MHY1485 for 12 hours, and observed that the level of mitophagy was decreased ( Figure 7E ). Similarly, after treatment, ROS levels were decreased ( Figure 7F ) while mitochondrial membrane potential ( Figure 7G ) and ATP levels ( Figure 7H ) were restored in DUSP6 knockdown cell line. There results indicate that the DUSP6 mediates mitophagy through mTOR pathway in bladder cancer.