Abstract Background Urothelial bladder cancer (UBC) is a highly heterogeneous malignancy with poor prognosis in muscle-invasive and high-grade subtypes. Epithelial-mesenchymal transition (EMT) drives tumor aggressiveness, yet its molecular mechanisms in UBC remain unclear. BAI1 associated protein 2 (BAIAP2) has been linked to cancer progression but remains unexplored in UBC. This study investigates the expression, functional role, and regulatory mechanisms of BAIAP2 in UBC, focusing on its contribution to tumor aggressiveness. Methods This study investigated the role of BAIAP2 in UBC using single-cell data analysis, bioinformatics, and functional assays. BAIAP2 expression was analyzed across UBC sub-populations, stages, and molecular subtypes via immunohistochemistry and quantitative methods. Transwell migration, invasion, and wound-healing assays were used to assess the impact of BAIAP2 knockdown and overexpression on cell behavior. EMT-like changes were examined through immunofluorescence and bright-field imaging. The roles of BAIAP2 in regulation of EMT pathways and its interaction with the transcription factor RELA were validated by Western blot analysis. Enrichment analysis of TCGA-BLCA datasets identified associated gene ontology terms and KEGG pathways. Results BAIAP2 was overexpressed in UBC, particularly in muscle-invasive and high-grade subtypes, and correlated with poor prognosis. Functional assays showed BAIAP2 promoted migration, invasion, and EMT-like changes, while its knockdown suppressed these behaviors. Bioinformatics analysis linked BAIAP2 to the transcription factor RELA, with RELA knockdown reducing BAIAP2 expression. Enrichment analysis implicated BAIAP2 in cytoskeletal reorganization and tumor progression, highlighting its role in UBC aggressiveness and potential for further therapeutic investigation. Conclusions BAIAP2 was highly expressed in muscle-invasive and high-grade tumors and was associated with poor prognosis. It promoted metastasis and EMT through activation of cytoskeletal remodeling. These findings identified BAIAP2 as a promising biomarker and a potential therapeutic target for the aggressive UBC. Supplementary Information The online version contains supplementary material available at 10.1186/s12885-025-14470-9. Keywords: Urothelial bladder cancer (UBC), BAI1 associated protein 2 (BAIAP2), Epithelial-mesenchymal transition (EMT), Progression Introduction Urothelial bladder cancer (UBC) is one of the most prevalent malignant tumors of the urinary system, characterized by poor prognosis and significant heterogeneity [[32]1]. Approximately 75% of UBC cases are diagnosed as non-muscle invasive bladder cancer (NMIBC), and about 10–20% of NMIBC will progress to muscle-invasive bladder cancer (MIBC) [[33]2]. The prognosis of UBC remains unsatisfactory due to high recurrence and metastasis rates [[34]3]. Currently, the effectiveness of therapeutic approaches for UBC is limited [[35]4]. Therefore, identifying specific prognostic markers is crucial for developing novel therapeutic strategies. In UBC, metastasis often involves lymph node colonization, with the bone (approximately 24%) and urinary system (approximately 23%) being the most common sites of secondary tumor development. Other less frequent sites of metastasis include the lung, liver, and brain [[36]5]. Metastasis is a complex, multifaceted process that encompasses several critical events, such as basement membrane invasion and cellular migration. Recent advancements in research have significantly enhanced understanding of the molecular and cellular mechanisms underlying metastatic progression [[37]6–[38]8]. Despite these advancements, the comprehension of metastasis remains incomplete. BAIAP2, also known as IRSp53, plays a pivotal role in orchestrating essential cellular processes such as migration and invasion by regulating actin and filopodium formation [[39]9–[40]12]. IRSp53 functions as an adaptor protein, bridging Rho-family small GTPases and actin cytoskeleton reorganizing proteins [[41]13]. Notably, studies have demonstrated that mDia associates with IRSp53 in a Rho-GTP-dependent manner [[42]11, [43]14]. Additionally, IRSp53 interacts with a multitude of actin regulators, including Rac1 [[44]15, [45]16], shank1 [[46]17], VASP [[47]18], N-WASP [[48]19, [49]20], WAVE2 [[50]11, [51]21], Cdc42 [[52]15], and Eps8 [[53]13, [54]22]. IRSp53’s involvement in actin dynamics is critical for its role in cellular motility and structural integrity. Extensive researches have highlighted the significant association between BAIAP2 and the progression of various cancers, including breast cancer, colorectal cancer, and hepatocellular carcinoma [[55]23–[56]30]. These studies suggest that BAIAP2 may influence tumor cell behavior by affecting cytoskeletal dynamics and cell signaling pathways. For instance, the interaction between IRSp53 and Rac1 is crucial for the regulation of lamellipodia formation, which is essential for cell migration and invasion in cancer metastasis [[57]15, [58]16]. Despite these insights, the specific role of BAIAP2 in the development and progression of UBC remains inadequately understood. Further investigations are required to elucidate the molecular mechanisms by which BAIAP2 contributes to UBC pathogenesis. In this study, we found that BAIAP2 was upregulated in UBC tissues, especially in the Basal/Squamous subtype, and high expression levels were associated with poor prognosis. In vitro experiments showed that BAIAP2 influenced key cellular processes like migration and invasion. These findings suggest that BAIAP2 can serve as a valuable prognostic marker and a potential therapeutic target for UBC. Methods Clinical data acquisition This study involved the collection of 107 formalin-fixed paraffin-embedded (FFPE) UBC samples from the Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University (WMU). Prior to sample collection, written informed consent was obtained from all participating patients, ensuring their voluntary participation and understanding of the study’s purpose, procedures, and potential risks. The study protocol was reviewed and approved by the Research Ethics Committee of the Second Affiliated Hospital and Yuying Children’s Hospital of WMU (Approval No. 2021-K-101-01). The ethical approval process adhered to the principles outlined in the Declaration of Helsinki, and all patient data were anonymized to protect confidentiality. The collection of FFPE samples has been described in a previous study [[59]31]. Immunohistochemistry (IHC) FFPE tissue sections (3 µm) were processed on the Ventana Benchmark Ultra automated staining system (Roche Diagnostics, USA). Briefly, tissue sections were deparaffinized and rehydrated prior to antigen retrieval using Cell Conditioner 1 (prediluted; pH 8.0) at 37 °C for 30 min. After blocking endogenous peroxidase activity, sections were incubated with the primary antibody against IRSp53/BAIAP2 (dilution 1:400; Abcam, USA) for 32 min at 37 °C. The detection procedure was then performed using the UV HRP UNIV MULT (secondary antibody) for 8 min, followed by 8 min incubations with UV DAB and UV DAB H[2]O[2]. Slides were then counterstained with hematoxylin, dehydrated, and coverslipped. BAIAP2 (IRSp53) expression was quantitatively evaluated using the H-score method, based on staining intensity (0: negative; 1: weak; 2: moderate; 3: strong) and the percentage of positive cells. Two independent pathologists, who were blinded to the clinical data, evaluated the slides using standardized procedures to ensure accuracy and reproducibility. Discrepancies in scoring were resolved through consensus discussion. Cell culture In this study, we utilized two human UBC cell lines: T24 and UM-UC-3. T24 cells were cultured in RPMI-1640 medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Biological Industries, Israel) and 1% penicillin/streptomycin (Gibco, USA). UM-UC-3 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, USA) under the same conditions, with 10% FBS and 1% penicillin/streptomycin. Both cell lines were maintained in a humidified incubator at 37 °C with 5% CO[2] to ensure optimal growth conditions. The culture medium was refreshed every 2–3 days, and cells were passaged at 80–90% confluence using 0.25% trypsin-EDTA (Gibco, USA). All experiments were performed with cells at passages 5–20 to ensure consistency and reproducibility. RNA interference (RNAi) For RNAi experiments, T24 and UM-UC-3 cells were seeded into 6-well plates at a density of 2 × 10^5 cells per well and allowed to adhere overnight. The cells were then transfected with small interfering RNAs (siRNAs) using Lipofectamine RNAiMax (Life Technologies, USA) according to the manufacturer’s protocol. The siRNAs targeting BAIAP2 and the corresponding negative control siRNAs were purchased from GenePharma (Shanghai, China). Similarly, siRNAs targeting RELA and their negative controls were obtained from IGEbio (Guangzhou, China). Transfection complexes were prepared by diluting siRNAs and Lipofectamine RNAiMax in Opti-MEM medium (Gibco, USA) and incubating for 20 min at room temperature. The complexes were added to the cells, which were then incubated at 37 °C in a 5% CO[2] humidified incubator. After 72 h post-transfection, the cells were harvested for subsequent analysis. Western blot analysis was performed to assess the knockdown efficiency and the impact on protein expression. The specific target sequences used for silencing control (siNC), si-BAIAP2-1, and si-BAIAP2-2 were as follows: siNC:5’-UUCUCCGAACGUGUCACGUTT-3’; si-BAIAP2-1: 5’-GGAAGAAAUGCUGAAGUCUTT-3’; si-BAIAP2-2: 5’-GCUGAAGAAAUACCAGACUTT-3’. For RELA, the target sequences for siNC, si-RELA-1, and si-RELA-2 were: siNC: 5’-UUCUCCGAACGUGUCACGUdTdT-3’; si-RELA-1: 5’-GGACAUAUGAGACCUUCAAdTdT-3’; si-RELA-2: 5’-GCCCUAUCCCUUUACGUCAdTdT-3’. Plasmid transfection For the plasmid transfection experiments, pCDH/BAIAP2 and pCDH/Vector plasmids were procured from IGE Biotechnology (Guangzhou, China). The BAIAP2 gene sequence was cloned into the pCDH vector following the manufacturer’s protocol to ensure accurate insertion and expression. The plasmid DNA and X-tremeGENE HP reagent (Sigma, USA) were mixed in serum-free medium to form DNA-reagent complexes, which were then added to the cells. After 72 h of incubation, the transfected cells were harvested for subsequent analysis. Immunofluorescence staining At 48 h post-transfection, cells were fixed with 4% paraformaldehyde for 15 min at room temperature and washed three times with phosphate-buffered saline (PBS). Cells were then incubated with rhodamine-labeled phalloidin (dilution 1:80; ABclonal Technology, China) to visualize F-actin. Following incubation, cells were washed three times with PBS and counterstained with DAPI (Beyotime, China) to label nuclei. After DAPI staining, cells were washed three times with PBS to remove unbound dye. Images were captured using an Olympus IX71 inverted fluorescence microscope (Japan) with a 40× objective. Cell morphology assay To assess the morphological changes induced by BAIAP2 overexpression, T24 and UM-UC-3 cells were transfected with either a BAIAP2 overexpression plasmid or an empty vector (control). Cells were seeded into 6-well plates at a density of 2 × 10^5 cells per well and allowed to adhere overnight. After 48 h of transfection, bright-field images were captured using an Olympus IX71 inverted fluorescence microscope (Japan) with a 20× objective. Cell migration and invasion assays To evaluate cell migration and invasion, transwell assays were performed using transwell chambers with an 8.0 μm pore size (Millipore, Germany). T24 and UM-UC-3 cells were suspended in serum-free medium and seeded into the upper chambers. For invasion assays, the upper chambers were pre-coated with Matrigel matrix (Corning, USA) to simulate the extracellular matrix, while migration assays were conducted without Matrigel. The lower chambers were filled with medium containing 10% FBS, which served as a chemoattractant to promote cell movement. After 24 h of incubation at 37 ℃ in a humidified atmosphere with 5% CO[2], the non-migrated or non-invaded cells remaining on the upper surface of the membrane were gently removed using a cotton swab. The transwell membranes were then fixed with 4% paraformaldehyde (Servicebio, China) for 20 min to preserve the cellular structures. Following fixation, the cells that had migrated or invaded to the lower surface of the membrane were stained with 0.1% crystal violet (Servicebio, China) for 15 min. This staining allowed for the visualization and quantification of the cells under an orthographic-view microscope (Nikon, Japan). Western blot assay For the Western blot analysis, cells were lysed using radioimmunoprecipitation assay (RIPA) lysis buffer (Beyotime, China) to extract total protein, which was then quantified using a BCA assay (Thermo Fisher Scientific, USA). Proteins were separated on 10% polyacrylamide gels (Epizyme, China) and transferred to PVDF membranes (Merck Millipore, Germany). The membranes were blocked with 5% nonfat milk in TBST and incubated with primary antibodies overnight at 4 ℃. The primary antibodies used were anti-IRSp53 (dilution 1:1000; Abcam, USA), anti-Vimentin (dilution 1:2000; CST, USA), anti-Claudin 1 (dilution 1:1000; CST, USA), anti-GAPDH (dilution 1:10000; Proteintech, USA). After washing, the membranes were treated with HRP-conjugated secondary antibody (CWBio, China) for 1 h. Finally, membranes were detected by enhanced chemiluminescence (Thermo Fisher Scientific, USA) using SmartChemi 910 plus (Sinsitech, China). Quantification of band signal intensities was performed using ImageJ. Single-cell RNA sequencing analysis The single-cell RNA sequencing (scRNA-seq) data for eight cases of primary bladder cancer, comprising two low-grade and six high-grade bladder urothelial carcinomas, were sourced [[60]32] from the DISCO scRNA repository ([61]https://www.immunesinglecell.org/repository), where the data were preprocessed and integrated. Quality control and data integration were executed using the platform’ s built-in fast integration tool [[62]33]. Subsequent analyses were performed utilizing the R programming language. Gene expression was visualized with UMAP plots, and violin plots were used to show the distribution of gene expression within each subpopulation. TCGA/BLCA bulk RNA-Seq data analysis Gene expression data were sourced from The Cancer Genome Atlas (TCGA) database ([63]https://tcga-data.nci.nih.gov/tcga/). For data analysis and visualization, we utilized custom R scripts developed in-house. Survival and correlation analyses were performed with R packages integrated into these scripts. For the TCGA cohort, the optimal cutoff for BAIAP2 expression stratification was determined using the ‘surv_cutpoint’ function from the R package ‘survminer’, which employed maximally selected rank statistics to identify the threshold (10.5 in this case) that best separates survival outcomes (low: ≤10.5; high: >10.5). Statistical analysis for experimentation data In our cohort, BAIAP2 expression groups were defined based on biological detection thresholds, with samples showing zero expression intensity of BAIAP2 classified as the low-expression group and those with any detectable expression (intensity > 0) as the high-expression group. Statistical significance of experimental data was determined using the t-test or Wilcoxon rank-sum test, with a P-value less than 0.05 being considered statistically significant. Graphical representations were created using GraphPad Prism version 8, ensuring clear and impactful visualization of the results. Each experiment was performed in triplicate to ensure reproducibility. Results BAIAP2 expression in the sub-populations by scRNA-seq in UBC and its clinical significance We analyzed single-cell RNA sequencing (scRNA-seq) data from eight UBC samples, sourced from the DISCO scRNA repository, to explore gene expression patterns across various cellular subpopulations. Based on gene expression markers, cells were classified into 16 distinct types (Fig. [64]1A). Notably, BAIAP2 was highly expressed in Basal Cells, Activated Dendritic Cells (DCs), and Suprabasal Epithelial Cells (Fig. [65]1B-D). Furthermore, the expression levels of BAIAP2 were significantly associated with different stages of UBC, showing particularly higher levels in patients with stages III and IV compared to those with stage II (P = 0.005, Fig. [66]1E). Additionally, the forest plot illustrated that the hazard ratios (HR) for various factors were associated with overall survival in TCGA-BLCA cohort, highlighting that BAIAP2 expression, age, and clinical stage were independent prognostic factors for UBC (Fig. [67]1F). To further investigate the relationship between BAIAP2 expression and clinicopathological characteristics, we analyzed our own cohort of 107 UBC patients. Our findings indicated that higher BAIAP2 expression was significantly associated with higher histological grade (P < 0.001), advanced T stage (P = 0.002) and muscle invasion (P = 0.002) (Table [68]1). Kaplan-Meier survival analysis demonstrated that patients with higher BAIAP2 expression had significantly shorter overall survival compared to those with lower expression levels (Fig. [69]1G). Further survival analysis using TCGA-BLCA data confirmed that UBC patients with high BAIAP2 expression exhibited poorer overall survival rates than those with low expression (Fig. [70]1H). Fig. 1. [71]Fig. 1 [72]Open in a new tab BAIAP2 expression in sub-populations of UBC tissues. (A) Clustering analysis of primary cell types. (B,C) BAIAP2 expression in the sub-populations, visualized in UMAP. (D) Violin plot shows expression level of BAIAP2 across major cell types. (E) Violin plots of BAIAP2 expression levels across three cancer stages: Stage II, Stage III, and Stage IV. (F) The forest plot illustrates the HR for various factors associated with BAIAP2 (G) Kaplan-Meier overall survival curves according to BAIAP2 expression levels in our in-house cohort of 107 UBC patients. The number at risk at each time point is shown below the survival curves (Note: Three cases were excluded due to missing event status). (H) High level of BAIAP2 predicted poor prognosis in TCGA-BLCA cohort. Log-rank P value < 0.05 was considered as statistically significant Table 1. Correlation between BAIAP2 expression levels and clinicopathological variables in urothelial carcinoma Variable Number (%) OR (95%CI) P-value Intensity_Low Intensity_High Gender  Male 32(78) 54(82)  Female 9(22) 12(18) 0.8(0.3–2.4) 0.627 Age (years)  ≤ 60 7(17) 15(23)  >60 34(83) 51(77) 0.7(0.2–2.1) 0.624 Grade  Low 34(83) 12(18)  High 7(17) 54(82) 21.9(7.1–70.7) <0.001 Muscle Invasion  Missing 7(17) 1(1)  NMIBC 29(71) 34(52)  MIBC 5(12) 31(47) 5.3(1.7–19.4) 0.002 T stage  Missing 7(17) 1(1)  T1/Ta 29(71) 34(52)  T2-T4 5(12) 31(47) 5.3(1.7–19.4) 0.002 [73]Open in a new tab Elevated BAIAP2 expression in aggressive molecular subtypes of UBC To gain deeper insights into the role of BAIAP2 in UBC, we analyzed its protein expression across a cohort of UBC clinical samples, scoring the expression levels from 0 to 3 (Fig. [74]2A). Our analysis revealed that BAIAP2 expression was significantly higher in tumors compared to adjacent non-tumor tissues (Fig. [75]2B). Furthermore, BAIAP2 expression was markedly elevated in MIBC compared to NMIBC (Fig. [76]2C and D), and higher in high-grade tumors compared to low-grade tumors (Fig. [77]2E and F). Fig. 2. [78]Fig. 2 [79]Open in a new tab The expression of BAIAP2 in UBC tissues. (A) Representative images showing the expression of BAIAP2 in 107 UBC tissues detected by IHC. Score from 0 to 3. (B) Representative images showing the expression of BAIAP2 in para-tumor and tumor tissues detected by IHC. (C) H-score of BAIAP2 in NMIBC (n = 63) and MIBC (n = 36) tissues (Note: Eight samples were excluded due to missing data). (D) Representative images showing the expression of BAIAP2 in NMIBC and MIBC tissues detected by IHC. (E) H-score of BAIAP2 in Low-grade (n = 46) and High-grade (n = 61) tissues. (F) Representative images showing the expression of BAIAP2 in Low-grade and High-grade tissues detected by IHC. Wilcox tests were used to evaluate the statistical significance of differences between two groups. ***P < 0.001 Basal/Squamous subtypes of bladder cancer are characterized by aggressive behavior and poor prognosis, making them a critical focus of research. To investigate the association of BAIAP2 expression with Basal/Squamous features and molecular subtypes in UBC, we performed a comprehensive analysis using bioinformatics and experimental approaches. IHC analysis revealed high BAIAP2 expression in Basal/Squamous subtypes (Fig. [80]3A). A strong positive correlation was observed between BAIAP2 expression and Basal/Squamous markers (Fig. [81]3B, Supplementary Figure [82]S1A). Furthermore, tumors with high BAIAP2 expression exhibited significantly higher levels of Basal/Squamous markers, including KRT5, KRT14 and S100A10, compared to those with low BAIAP2 expression (Supplementary Figure [83]S1B). Additionally, analysis of molecular subtypes showed that BAIAP2 expression was significantly elevated in Basal/Squamous subtypes compared to luminal and other subtypes (Fig. [84]3C). These findings suggest that BAIAP2 is closely associated with Basal/Squamous features and may play a role in defining the molecular heterogeneity of bladder cancer. Fig. 3. [85]Fig. 3 [86]Open in a new tab Association of BAIAP2 expression with Basal/Squamous features and molecular subtypes in UBC. (A) Representative images showing Basal/Squamous differentiation in tumors with high BAIAP2 expression. (B) Scatter plots to illustrate the correlation of BAIAP2 expression with Basal/Squamous cell markers. (C) Expression levels of BAIAP2 in different molecular subtypes of UBC (TCGA-BLCA) BAIAP2 silencing inhibits bladder cancer cell migration and invasion To investigate the biological functions of BAIAP2 in the aggressive behavior of UBC cells, we performed BAIAP2 knockdown in two UBC cell lines using siRNA-mediated gene silencing. Western blot analysis confirmed a significant reduction in BAIAP2 protein levels following siRNA transfection, validating the efficiency of BAIAP2 knockdown (Fig. [87]4A). Fig. 4. [88]Fig. 4 [89]Open in a new tab Knockdown of BAIAP2 inhibited migration and invasion of UBC cells. (A) The protein abundance of BAIAP2 in UBC cells measured by western blot analysis. (B) Representative images and quantity of transwell migration and invasion assays. (C) Representative images and quantity of wound healing assays. (Data are presented as mean ± SD, n = 3. t-test, **P < 0.01) To assess the impact of BAIAP2 silencing on cell migration and invasion, we conducted transwell and scratch wound-healing assays. Transwell invasion assays revealed a significant decrease in the number of BAIAP2-silenced cells penetrating the Matrigel-coated membrane, suggesting a suppression of invasive potential (Fig. [90]4B). Similarly, in the scratch wound-healing assay, BAIAP2-depleted cells exhibited markedly reduced wound closure compared to control cells, indicating impaired migratory capacity (Fig. [91]4C). These findings demonstrate that BAIAP2 silencing effectively inhibits the migration, invasion, and wound-healing capabilities of UBC cells, highlighting its critical role in promoting the aggressive behavior of UBC. BAIAP2 overexpression promotes migration and invasion in UBC cells Having demonstrated that BAIAP2 silencing inhibits UBC cell migration and invasion, we next sought to investigate whether BAIAP2 overexpression could enhance these aggressive behaviors. BAIAP2 expression was significantly up-regulated after transfection with the pCDH-BAIAP2 plasmid, as confirmed by Western blot analysis (Fig. [92]5A). Functional assays further demonstrated that BAIAP2 overexpression enhanced cellular migration, invasion, and wound-healing capabilities. Transwell invasion assays revealed a significant increase in the number of BAIAP2-overexpressing cells penetrating the Matrigel-coated membrane, indicating heightened invasive potential (Fig. [93]5B). Similarly, scratch wound-healing assays showed accelerated wound closure in BAIAP2-overexpressing cells compared to controls, suggesting enhanced migratory capacity (Fig. [94]5C). Fig. 5. [95]Fig. 5 [96]Open in a new tab Over expression of BAIAP2 promoted migration and invasion of UBC cells in vitro. (A) Western blot analysis showed overexpression of BAIAP2 in T24 and UM-UC-3 cells. (B) Representative images and quantity of transwell migration and invasion assays. (C) Representative images and quantity of wound healing assays. (Data are presented as mean ± SD, n = 3. t-test, *P < 0.05, **P < 0.01) These findings collectively indicate that BAIAP2 overexpression promotes the migratory and invasive properties of bladder cancer cells, further supporting its role in driving tumor progression. Overexpression of BAIAP2 induces epithelial-mesenchymal transition phenotypes To elucidate the underlying molecular mechanisms of BAIAP2 in regulation of bladder cancer progression and metastasis, we performed a series of in vitro experiments and bioinformatics analyses. Immunofluorescence staining revealed significant cytoskeletal reorganization in T24 and UM-UC-3 cells overexpressing BAIAP2, including the formation of elongated, spindle-shaped cells and filopodia, as indicated by the white arrows (Fig. [97]6A). Similarly, bright-field imaging confirmed that BAIAP2-overexpressing cells exhibited EMT-like morphological changes, such as cell elongation and loss of cell-cell contact, compared to control cells (Fig. [98]6B). Fig. 6. [99]Fig. 6 [100]Open in a new tab BAIAP2 promoted EMT-like changes of UBC cells in vitro. (A) T24 and UM-UC-3 cells transfected with BAIAP2 or control plasmid were stained with phalloidin (F-actin, red) and DAPI (nuclei, blue). White arrows indicate morphological changes. Scale bar: 20 μm. (B) Representative bright-field images of T24 and UM-UC-3 cells overexpressing BAIAP2 or transfected with control plasmid. Scale bar: 50 µm. White arrows indicate EMT-like changes. (C) Effects of BAIAP2 knockdown on EMT marker expression. (Top) Representative Western blots showing Vimentin and Claudin1 protein levels in T24 and UM-UC3 cells. (Bottom) Quantitative analysis of Vimentin and Claudin1 expression normalized to GAPDH (mean ± SD, n = 3. t-test, *P < 0.05, *** P < 0.001). (D) Violin plot showing the distribution of EMT scores in BAIAP2 high-expression and low-expression groups based on bioinformatics analysis of publicly available bladder cancer datasets (TCGA-BLCA). (E) Violin plot showing the distribution of EMT scores in different subtypes of UBC Then, we performed Western blot analysis of EMT markers in BAIAP2-depleted T24 and UM-UC-3 cells. BAIAP2 knockdown induced significant molecular changes consistent with EMT reversal, including downregulation of Vimentin (mesenchymal marker) and upregulation of Claudin-1 (epithelial marker). In UM-UC-3 cells, Vimentin expression was reduced to 0.58 ± 0.32-fold of control levels (P < 0.05), whereas Claudin-1 expression increased 6.09 ± 1.68-fold (P < 0.001). Similarly, in T24 cells transfected with si-BAIAP2-2, Vimentin expression decreased significantly to 0.57 ± 0.13-fold (P < 0.001), accompanied by a 1.57 ± 0.1-fold increase in Claudin-1 (P < 0.05). Although T24 cells transfected with si-BAIAP2-1 showed a more modest but still significant reduction in vimentin expression (0.79 ± 0.06-fold, P < 0.05), the increase in Claudin-1 (1.25 ± 0.22-fold) did not reach statistical significance (P = 0.07), likely due to the slightly lower knockdown efficiency of this siRNA construct (Fig. [101]6C). Bioinformatics analysis of publicly available bladder cancer datasets further supported these findings. Violin plots showed that BAIAP2 high-expression groups exhibited significantly higher EMT scores compared to low-expression groups (Fig. [102]6D). Additionally, EMT scores varied across different molecular subtypes of bladder cancer, with Basal/Squamous subtypes showing the highest scores (Fig. [103]6E). These results collectively indicate that BAIAP2 promotes EMT-like changes in bladder cancer cells, both in vitro and in clinical datasets, and may play a critical role in driving tumor progression and subtype-specific behaviors. Enrichment analysis of BAIAP2 gene co-expression network in UBC To explore further the roles of BAIAP2, we conducted a co-expression network analysis to identify genes that were significantly correlated with the gene expression levels of BAIAP2 gene. As shown in the volcano plot, NIBAN2, ANXA2, LRRC8A, RHOD, MAP7D1, and SLC9A3R1 exhibited the strongest correlation with BAIAP2 expression levels in the TCGA-BLCA data (Fig. [104]7A). Fig. 7. [105]Fig. 7 [106]Open in a new tab Enrichment analysis of BAIAP2 gene co-expression network in UBC. (A) The volcano plot showed co-expression genes associated with BAIAP2 expression in the TCGA-BLCA datasets. (B) Enrichment analysis of GO terms for co-expression genes which were positively correlated with BAIAP2. (C) GO terms for co-expression genes which were negatively correlated with BAIAP2. (D) Enrichment analysis of KEGG terms for co-expression genes which were correlated with BAIAP2 Among the top correlated genes, RHOD, one of the small GTPase superfamily members, is involved in endosome dynamics, and participates in the regulation of filopodia formation and actin filament bundling. In addition, LRRC8A is a subunit of the volume-regulated anion channel (VRAC, also named VSOAC channel) that mediates efflux of amino acids, and is required for the uptake of the cisplatin [[107]34]. The interactions between these top correlated genes and BAIAP2 in contribution to the progression of bladder cancer warrants further investigation. Furthermore, Gene Ontology (GO) analyses were conducted to evaluate the top 200 genes co-expressed with BAIAP2, including those positively and negatively correlated. We found that co-expression of BAIAP2 was positively associated with multiple biological processes, such as proteasome complex and protein binding (Fig. [108]7B). Conversely, it was negatively correlated with RNA binding and RNA splicing (Fig. [109]7C). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis identified that the co-expressed genes were positively correlated with BAIAP2 expression level were primarily involved in ERBB and WNT signaling pathways (Fig. [110]7D), which were essential for cell proliferation and differentiation. Transcription factors may influence the transcription of BAIAP2 To identify the up-stream gene regulatory networks of BAIAP2, we conducted an analysis of TF-target pairs for BAIAP2 by using GRNdb tools ([111]http://www.grndb.com/) [[112]35], we found that the transcription factor RELA, KLF5, and TFAP2C might bind to the regulation motifs of BAIAP2 to regulate gene expression (Fig. [113]8A). Moreover, gene co-expression analysis revealed that BAIAP2 was correlated with the expression of RELA (Fig. [114]8B), suggesting that BAIAP2 may enhance the transcription of BAIAP2 through its TF-target interaction with RELA. Fig. 8. [115]Fig. 8 [116]Open in a new tab Association of BAIAP2 with the transcription factor RELA and its functional validation. (A) RELA is one top associated transcription factor for BAIAP2. (B) Correlation between RELA and BAIAP2 expression in BLCA, based on data from the TCGA-BLCA database. (C) The expression of BAIAP2 was down-regulated in T24 and UM-UC-3 cells after siRELAs transfection To validate the findings in bioinformatic analysis, we transfected T24 and UM-UC-3 cells with siRNAs targeting RELA (siRELAs) and observed a significant downregulation of BAIAP2 expression (Fig. [117]8C), highlighting the regulatory relationship between BAIAP2 and RELA in tumor progression. However, the detailed interactions between RELA, KLF5, and TFAP2C and the cis-regulation elements of BAIAP2 warrant further investigations. Discussion In this study, we identified that BAIAP2 was consistently upregulated in UBC, correlating with poor prognosis. Notably, higher BAIAP2 expression was significantly associated with higher histological grade and advanced T stages, suggesting that it might play a critical role in the progression of UBC. BAIAP2 encodes IRSp53, a scaffolding protein that bundles actin filaments and interacts with the small GTPase Rac [[118]36]. Previous studies have shown that BAIAP2 is expressed in filopodia, interacts with Cdc42 and Rac, and is critical for axon guidance and neurite outgrowth [[119]9]. Our results support the notion that BAIAP2 enhances the migration, invasion, and wound-healing capacities of UBC cells. Further experimental studies are warranted to determine whether BAIAP2 directly binds to Rac to promote tumor progression. In UBC, few studies have systematically correlated molecular subtype profiles with tumor metastasis. One novel aspect of our study is the observed correlation of BAIAP2 with the Basal/Squamous subtypes of UBC, which are known for their aggressive nature associated with immune cell infiltration and fibroblast enrichment. Although we found that BAIAP2 was positively correlated with various immune cell markers, the roles of BAIAP2 in immune regulation warrant further investigation. Furthermore, our pathway enrichment analysis indicated that BAIAP2-associated tumor progression pathways were mainly enriched in differentiation. Meanwhile, Gene Set Enrichment Analysis (GSEA) of BAIAP2 co-expression network suggested that the activation of multiple hallmark and Wiki pathways might drive disease progression, with TNF-α and TGF-β involved in inflammation and differentiation, and epithelial to mesenchymal transition playing critical role in metastasis (Supplementary Figure [120]S2A, [121]2B). These findings suggested that BAIAP2 was critical for tumor differentiation and progression. However, the mechanisms through which BAIAP2 regulates tumor cell progression remain incompletely understood in our study. We found that BAIAP2 was associated with transcription factor RELA. Knockdown of RELA in bladder cancer cells resulted in a significant decrease in BAIAP2 expression, suggesting a regulatory role of RELA in BAIAP2 expression and tumor progression. Further in-depth investigations into the molecular mechanisms by which BAIAP2 influences these pathways can provide valuable insights into its potential as a therapeutic target in UBC. Despite the novel insights of this study, several limitations should be acknowledged. First, the cohort size, particularly for subgroup analyses across molecular subtypes and advanced clinical stages, may limit the generalizability of the conclusions, although partial validation was achieved using public datasets. Second, while mechanistic explorations were conducted in vitro, the lack of in vivo validation through models such as patient-derived xenograft or orthotopic bladder cancer models leaves the physiological relevance of these findings undetermined. Third, although we identified RELA as a transcriptional regulator of BAIAP2, the precise molecular mechanism remains unresolved and warrants further investigation. Finally, clinical translation of BAIAP2-targeted strategies faces many challenges, such as potential off-target effects and subtype-specific therapeutic efficacy disparities. Overall, our research highlights the critical function of BAIAP2 in UBC. Increased BAIAP2 expression is associated with both worse prognostic outcomes and is associated with aggressive molecular subtypes. BAIAP2 was critical for tumor differentiation and enhanced the transcription of genes involved in tumor progression. All of these revealed that BAIAP2 might serve as a prognostic biomarker and a therapeutic target in UBC. Conclusions In conclusion, our study demonstrates that elevated BAIAP2 expression is significantly linked to the progression of UBC. BAIAP2 enhances the migratory and invasive capabilities of UBC cells, correlating with advanced clinical features and aggressive molecular subtypes. These findings suggest that BAIAP2 may serve as both a prognostic biomarker and a potential therapeutic target for UBC. Further studies are needed to elucidate the precise mechanisms and explore its clinical translational applications. Electronic supplementary material Below is the link to the electronic supplementary material. [122]Supplementary Material 1^ (990.7KB, pdf) [123]Supplementary Material 2^ (19.6MB, zip) [124]Supplementary Material 3^ (5.5MB, pdf) Acknowledgements