LncRNA ACTA2-AS1 acts as a bladder cancer prognostic biomarker and blocks malignant advance via miR-148b-3p/DNAJB4 Authors: Xing Chen, Qunfu Tang, Yingjie Wang, Jianbin Zhang, Lei Zhou Published: 2026 ABSTRACT BackgroundLong non-coding RNA (lncRNA) is crucial in bladder cancer (BCa) development. This research explores the clinical significance and underlying mechanisms of lncRNA ACTA2-AS1 in BCa.MethodsFrom the GEO database, differentially expressed lncRNAs (DElncRNAs) in BCa were identified, and tumor, along with adjacent non-tumor tissue samples from 132 patients, were collected. RT-qPCR was employed for the quantitative measurement of ACTA2-AS1, miR-148b-3p, and DNAJB4 expression. Chi-square analysis was employed to assess the association of ACTA2-AS1 with clinical features. Kaplan-Meier and Cox regression evaluated ACTA2-AS1’s prognostic value. CCK-8, Transwell, and flow cytometry assays evaluated cell proliferation, migration, invasion, apoptosis, and cell cycle arrest. DLR and RIP tests confirmed miR-148b-3p’s targeting of ACTA2-AS1 and DNAJB4.ResultsACTA2-AS1 showed significant downregulation in four GEO databases. Moreover, both ACTA2-AS1 and DNAJB4 were notably decreased in BCa tumor tissues, while miR-148b-3p increased significantly. Low ACTA2-AS1 expression correlates with poor tumor differentiation, positive lymph node metastasis, and TNM stage III-IV. Moreover, patients with low ACTA2-AS1 expression are associated with a significantly unfavorable prognosis. Mechanistically, miR-148b-3p targets ACTA2-AS1 and DNAJB4. The suppression impacts of ACTA2-AS1 on BCa cell proliferation, migration, and invasion, and its boost of apoptosis and cell cycle arrest were greatly reversed by miR-148b-3p.ConclusionLow ACTA2-AS1 expression is a potential biomarker for unfavorable prognosis in BCa patients. Furthermore, ACTA2-AS1 may counteract BCa’s malignant biological behaviors via targeting the miR-148b-3p/DNAJB4 axis. Background Bladder cancer (BCa) stands as one of the top ten most prevalent cancers globally and ranks second among malignancies of the urinary system [1, 2]. With high progression and recurrence rates [3], it often leads to poor prognosis, causing about 213,000 deaths annually [4]. Even after undergoing transurethral resection of BCa and receiving risk-stratified adjuvant treatment, the five-year survival outlook remains poor [5]. Furthermore, conventional prognosis evaluation methods mainly hinge on clinical and histopathological features, yet they have limitations in precisely forecasting individual patient outcomes [6]. Hence, discovering new prognostic biomarkers is vital for patient clinical care. Numerous studies show that long non-coding RNA (lncRNAs) have great promise in predicting tumor type, treatment efficacy, and prognosis. Their stability, high expression, wide distribution, easy detectability, and cost-efficiency have drawn significant clinical research interest. LncRNAs refer to RNA transcripts exceeding 200 nucleotides in length, which possess limited protein-coding potential [7, 8]. Abnormal lncRNA levels and functions are linked to diverse diseases, including cancer. For example, SBF2-AS1 demonstrates oncogenic effects in BCa [9]. DGUOK-AS1 also promoted the BCa cell proliferation and invasion [10]. Additionally, DLGAP1-AS2 has been pinpointed as a biomarker indicating a poor prognosis in BCa [6]. LncRNA smooth muscle α-actin 2-antisense 1 (ACTA2-AS1, also termed ZXF1 or UCOOOO1KFO.1), situated on human chromosome 10q23.31, has been previously implicated in prostate cancer progression [11], thyroid cancer metastasis [12], and ovarian cancer malignancy [13]. In our initial study, we utilized datasets GSE191036, GSE97239, GSE236932, and GSE190079 from the GEO database to pinpoint differentially expressed lncRNAs in BCa tumor and adjacent non-tumor tissues. Among the overlapping differentially expressed lncRNAs (DElncRNAs), ACTA2-AS1 showed significant downregulation, with fold changes of 9.33, 10.97, 6.14, and 98.37, respectively. Additionally, online bioinformatics databases like TCGA, StarBase, and GEPIA also revealed marked downregulation of ACTA2-AS1 in BCa tumor tissues. Yet, its clinical relevance and underlying mechanisms in BCa are still unclear. Given the research context, we hypothesized that ACTA2-AS1 may act as a tumor suppressor gene in BCa progression. To test this, we systematically analyzed ACTA2-AS1 expression differences in BCa tissues, adjacent non-cancerous tissues, and cell lines. Besides evaluating the five-year prognosis, we also studied its impact on BCa cell migration, apoptosis, and cell cycle arrest in cell culture. Combining bioinformatics with experimental methods, this study seeks to clarify ACTA2-AS1’s clinical significance and molecular mechanisms, offering new insights and strategies for targeted BCa therapy. Materials and methods Identification of differentially expressed lncRNAs (DElncRNAs) Independent RNA-seq datasets were sourced from the Gene Expression Omnibus (GEO). Specifically, GSE236932 includes 63 BCa tissue samples: 29 male tumor samples, 17 male peritumoral (normal) samples, 9 female tumor samples, and 8 female peritumoral samples. GSE191036 contains 4 normal peritumoral tissues and 4 BCa tumor tissues. GSE97239 has 3 normal tissues and BCa tissues at various TNM stages. GSE190079 comprises 5 BCa tumor tissues and 5 peritumoral tissues. Raw expression data from all GEO datasets were standardized with quantile normalization to reduce technical heterogeneity and ensure comparable expression levels among samples. Differential expression analysis was performed using the limma package in R software. DElncRNAs were screened with the cutoff criteria of log fold change (FC) > 2 and a P-value < 0.05. Based on the above differential expression results, volcano plots were plotted and visualized using the Bioinformatics online platform (WeiShengXin) [14] to intuitively exhibit the overall distribution of upregulated and downregulated lncRNAs. Bioinformatics analysis ACTA2-AS1 expression data in BCa were retrieved and validated from GEPIA, StarBase, and TCGA databases. Its target miRNAs, including miR-148b-3p, were predicted using the lncRNASNP2 online database. The target mRNAs of miR-148b-3p are predicted by Targetscan, ENCORI, Tarbase, microT_interactions, StarBase, microT-CDS, and miRDB databases. The lncLocatoR database predicts ACTA2-AS1’s subcellular location. Research subjects and tissue acquisition The Ethics Committee of Tongxiang First People’s Hospital approved this study, which followed the principles and guidelines of the Declaration of Helsinki. All patients provided written informed consent. Information about the samples is kept in the hospital database for research. Between June 2017 and February 2020, 132 patients who underwent BCa surgery at Tongxiang First People’s Hospital were enrolled. The inclusion criteria were as follows: 1) meeting BCa diagnostic criteria; 2) diagnosis confirmed via histopathological examination of tissue samples by three independent pathologists; 3) newly diagnosed without prior radiotherapy, chemotherapy, or immunotherapy. Exclusion criteria included: 1) concurrent malignancies; 2) infectious disease (e.g., HIV, syphilis, infectious liver disease); 3) history of cardiovascular disease or diabetes mellitus; 4) incomplete follow-up data. Cancerous and adjacent non—cancerous tissues (within 3 cm of the lesion) were collected during surgery and stored at -80℃ after washing. Patient demographic data, including age and gender, were recorded. Clinical information, such as tumor size, differentiation grade, TNM stage, lymph node metastasis status, and tumor subtype, was extracted from medical records. Patients underwent a five-year follow-up via hospitalizations, outpatient visits, and electronic communication (WeChat, email). Follow-up intervals were every three months for the first two years, then every six months thereafter. Survival status was tracked from surgery until death or five-year follow-up completion. Cell culture and transfection Human normal urinary tract epithelial cell line SV-HUC-1, along with human bladder carcinoma cell lines RT-4, T24, J82, UM-UC-3, 5637, and BFTC-905, were all procured from BeNa Culture Collection. SV-HUC-1 cells were grown in F-12 K medium containing 10% FBS and 1% penicillin-streptomycin. 5637 cells were cultured in RPMI1640 medium, whereas the other cell lines were maintained in DMEM medium. The pcDNA3.1 vector harboring ACTA2-AS1 (ACTA2-AS1 overexpression plasmid oe-ACTA2-AS1) and its negative control empty vector (oe-NC) were both procured from Genechem. miR-148b-3p mimic, miR-148b-3p inhibitor, mimic NC, and inhibitor NC were acquired from RioBio. Cells were seeded on 6-well plates. When they reached 70–80% confluence, the above-mentioned plasmid was mixed with oligonucleotides and transfected into cells using Lipofectamine 2000. The medium was replaced after 6 h, and cells were cultured for an additional 48 h for subsequent experiments. Real-time quantitative reverse transcription PCR (RT-qPCR) The TRIzol LS extraction kit was used to isolate total RNA from tissues and cells. Cell purity and concentration were measured with a NanoDrop 2000 spectrophotometer; an A260/A280 ratio of 1.8-2.0 indicated acceptable purity. Reverse transcription of 500 ng of mRNA and lncRNA or miRNA was carried out using the HiScript® III 1st Strand cDNA Synthesis kit or miRNA First Strand cDNA Synthesis kit. After combining cDNA, primers, either the PrimeScript™ miRNA RT-qPCR kit or PrimeScript™ RT-qPCR kit, and RNase-free ddH2O, the amplification reaction was run on a LightCycler 480 real-time fluorescent quantitative PCR system. GAPDH and cel-miR-39 served as internal reference genes for relative quantification. The 2−ΔΔCt method was applied to calculate the fold change in expression of ACTA2-AS1, DnaJ heat shock protein (Hsp40) family member B4 (DNAJB4), and miR-148b-3p. Cell viability assay Cellular viability was evaluated using the CCK-8 assay. A total of 5 × 103 cells/100 µL were seeded into a 96-well plate. Every 24 h, the well medium was replaced with a mixture of 10 µL CCK-8 and 90 µL DMEM. The plate was then incubated for 1 h, after which changes in cellular viability were measured at 450 nm. Caspase 3 activity detection The caspase 3 assay kit (colorimetric) was used to measure caspase 3 activity. First, 4 × 106 cells were washed with PBS and then lysed using 150 µL lysis buffer. The lysate was centrifuged at 1000× g for 1 min. Protein concentration was determined using the Bradford method. Next, 10 µL substrate, 40 µL assay buffer, and 50 µL cell sample were added to a 96-well plate, and incubated for 2 h, and the OD was measured at 450 nm using a microplate reader. Cell apoptosis assay Apoptosis was analyzed via flow cytometry. Briefly, 5 × 106 cells were harvested, resuspended in 200 µL binding buffer, and then incubated with 5 µL propidium iodide (PI) and 5 µL Annexin V-FITC under light-protected conditions for 15 min. After adding 300 µL binding buffer, apoptosis levels were measured using the BD FACSAria III system. Cell cycle assay Flow cytometry was used to assess how overexpressing ACTA2-AS1 and miR-148b-3p influences the cell cycle progression. The cells were seeded at 1.5 × 105 per well in 6-well plates, transfected with oe-ACTA2-AS1 or miR-148b-3p mimic, and cultured for 48 h. After harvesting, the cells were fixed overnight in 80% ethanol at -20℃. Before analysis, they were centrifuged, resuspended in 500 µL cold PBS with 5 µL Rnase A, and incubated for 30 min. Following another centrifugation, the cells were resuspended in PBS containing DAPI and Triton, incubated for 10 min, centrifugated again, and washed with PBS. Flow cytometry was then used to analyze cell cycle arrest in each group. Migration and invasion assays Cell migration was assessed using 8 μm Transwell chambers. Briefly, 1 × 104 cells were resuspended in serum-free DMEM and placed in the upper chamber, while the lower chamber was filled with DMEM containing 10% FBS. The chambers were placed in the incubator for a further 24 h. Methanol was then added to fix the cells, and a cotton swab was used to remove the non-migrated cells from the Transwell chamber. After that, the cells were stained with crystal violet for 20 min, rinsed with water, and counted under a microscope. Additionally, for invasion assays, pre-hydrate the matrix gel in the Transwell chambers by adding 50 µL of FBS-free DMEM before seeding cells and then follow the same procedure as described for the migration assays. Dual luciferase reporter assay The wild-type (WT) sequence of ACTA2-AS1 or DNAJB4 containing the predicted miR-148b-3p binding site was amplified and subcloned into the psiCHECK-2 dual-luciferase reporter vector (Promega, Madison, WI, USA) to generate the recombinant reporter plasmids ACTA2-AS1-WT or DNAJB4-WT. For the mutant constructs (ACTA2-AS1-MT and DNAJB4-MT), site-directed mutagenesis was performed to alter the seed region of the miR-148b-3p binding site (specifically, the complementary nucleotides of miR-148b-3p were replaced to disrupt base pairing). The mutagenesis was conducted using a site-directed mutagenesis kit according to the manufacturer’s instructions. For luciferase reporter assays, cells were seeded in 24-well plates at a density of 1 × 105 cells per well and cultured to 60–70% confluence. Cells were co-transfected with reporter plasmids (WT or MT) and miR-148b-3p mimic, or mimic NC using Lipofectamine 2000. After 48 h post-transfection, cells were lysed, and firefly and Renilla luciferase activities were measured using a dual luciferase reporter assay system (Promega). Relative luciferase activity was calculated as the ratio of firefly to Renilla luciferase activity, with Renilla activity serving as an internal control. RIP assay The Magna RIP kit is used for RIP analysis of T24 and UM-UC-3 cells. Briefly, cells are lysed with RIP lysis buffer, and the resulting lysate is incubated with magnetic beads conjugated to either anti-Argonauts 2 (Ago2) or anti-IgG antibodies. The enrichment levels of ACTA2-AS1 and miR-148b-3p in the co-precipitated RNA are then analyzed. Cell fractionation of RNA Cytoplasmic and nuclear RNA were extracted from BCa cells using a Norgen cytoplasmic/nuclear RNA purification kit per the manufacturer’s protocol. Subsequently, the abundance of ACTA2-AS1 was measured by RT-qPCR. Statistical analysis All statistical analyses, including data processing and visualization, were carried out via SPSS 23.0 and GraphPad Prism 9.0. Quantitative data are presented as mean ± SD, based on five biological replicates. Non-parametric variables underwent comparison using the chi-square test. Parametric variables were analyzed using a two-tailed Student’s t-test (two groups) or a one-way analysis of variance (ANOVA) (multiple groups). For post-hoc multiple comparisons following ANOVA, Tukey’s multiple comparison test was applied to control the family-wise error rate. Intergroup relationships were assessed using Pearson’s correlation coefficients. Five-year overall survival was evaluated using Kaplan-Meier analysis, and multivariable Cox regression was employed to compute hazard ratios (HR) and 95% confidence intervals (CI) for identifying independent prognostic factors. Statistical significance was defined as P < 0.05. Results LncRNA ACTA2-AS1 exhibits downregulation in BCa tumor tissues and is associated with clinical features By analyzing databases GSE191036, GSE190079, GSE97239, and GSE236932, we identified DElncRNAs in BCa. Among them, ACTA2-AS1 was found to be one of the eight overlapping DElncRNAs (Fig. 1A-E). Public databases, including TCGA, GEPIA, and Starbase, further validated the marked downregulation of ACTA2-AS1 in BCa tumor tissues (P < 0.05, Fig. 1F-H). In our study of 132 BCa patients, ACTA2-AS1 expression was notably lower in tumor tissues than in normal tissues (P < 0.001, Fig. 1I). Moreover, its levels declined progressively with advancing TNM stage (P < 0.001, Fig. 1J). Patients were divided into high- and low-expression groups based on the mean ACTA2-AS1 expression level (0.43 ± 0.18) in tumor tissues. Notably, low ACTA2-AS1 expression was associated with poorly differentiated BCa (P = 0.040), positive lymph node metastasis (P = 0.034), and advanced TNM stages (III-IV) (P = 0.023, Table 1). Finally, we also found that ACTA2-AS1 expression was markedly lower in BCa cell lines (5637, BFTC-905, J82, RT4, T24, UM-UC-3) compared to normal urinary tract epithelial cell line SV-HUC-1 cells, with the most significant reduction evident in T24 and UM-UC-3 cell lines (P < 0.05, Fig. 1K). Fig. 1LncRNA ACTA2-AS1 exhibits significant downregulation in BCa tumor tissue and cancer cells. A. A Venn diagram illustrates the overlapping DElncRNAs identified in BCa tumors across various GEO datasets. B-E. Volcano plots depicting the fold changes of ACTA2-AS1 expression in the GES191036, GSE190079, GSE97239, and GSE236932 datasets. F-H. The expression profiles of ACTA2-AS1 in BCa tumors and normal tissues were sourced from public databases, including TCGA, GEPIA, and StarBase. I. RT-qPCR quantified ACTA2-AS1 levels in BCa patients’ tumors and normal tissues. J. RT-qPCR assessed ACTA2-AS1 across BCa TNM stages. K. ACTA2-AS1 expression was compared between BCa and normal epithelial cell lines. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 LncRNA ACTA2-AS1 exhibits significant downregulation in BCa tumor tissue and cancer cells. A. A Venn diagram illustrates the overlapping DElncRNAs identified in BCa tumors across various GEO datasets. B-E. Volcano plots depicting the fold changes of ACTA2-AS1 expression in the GES191036, GSE190079, GSE97239, and GSE236932 datasets. F-H. The expression profiles of ACTA2-AS1 in BCa tumors and normal tissues were sourced from public databases, including TCGA, GEPIA, and StarBase. I. RT-qPCR quantified ACTA2-AS1 levels in BCa patients’ tumors and normal tissues. J. RT-qPCR assessed ACTA2-AS1 across BCa TNM stages. K. ACTA2-AS1 expression was compared between BCa and normal epithelial cell lines. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 Table 1Expressions level of lncRNA ACTA2-AS1 and its association with clinical pathological characteristics in bladder cancer patientsParametersCases No.(n = 132)ACTA2-AS1 expressionP valueHigh (n = 63)Low (n = 69)Age (years)0.389 < 55683533 ≥ 55642836Gender0.718 Female462323 Male864046Tumor size (cm)0.600 < 3593029 ≥ 3733340Subtype0.057 Papillary703931 Non-papillary622438Differentiation0.040 Well-moderate914942 Poor411427Lymph node metastasis0.034 Negative955144 Positive371225TNM stage0.023 I-II744232 III-IV582137 Table 1Expressions level of lncRNA ACTA2-AS1 and its association with clinical pathological characteristics in bladder cancer patientsParametersCases No.(n = 132)ACTA2-AS1 expressionP valueHigh (n = 63)Low (n = 69)Age (years)0.389 < 55683533 ≥ 55642836Gender0.718 Female462323 Male864046Tumor size (cm)0.600 < 3593029 ≥ 3733340Subtype0.057 Papillary703931 Non-papillary622438Differentiation0.040 Well-moderate914942 Poor411427Lymph node metastasis0.034 Negative955144 Positive371225TNM stage0.023 I-II744232 III-IV582137 Expressions level of lncRNA ACTA2-AS1 and its association with clinical pathological characteristics in bladder cancer patients Reduced expression of lncRNA ACTA2-AS1 may serve as a potential biomarker for unfavorable prognosis in BCa patients Over the 5-year follow-up, 30 patients (22.73% of the cohort) died. A comparison of ACTA2-AS1 expressions between patients with favorable and unfavorable prognoses showed significantly lower levels in those with unfavorable prognosis (P < 0.001, Fig. 2A). What’s more, the Kaplan-Meier curve indicated that patients with low ACTA2-AS1 expression had a worse prognosis (log-rank P = 0.009, Fig. 2B). Cox regression analysis identified ACTA2-AS1 (HR = 0.396, 95%CI: 0.171–0.920, P = 0.031) along with TNM staging (HR = 2.723, 95%CI: 1.064–6.964, P = 0.037) and lymph node metastasis (HR = 2.262 95%CI: 1.020–5.016, P = 0.044), as potential risk factors for an unfavorable prognosis in BCa (P < 0.05, Fig. 2C). Fig. 2Low expression of ACTA2-AS1 is associated with an unfavorable prognosis in BCa patients. A Comparison of ACTA2-AS1 expression between BCa patients with favorable and unfavorable prognoses. B Kaplan-Meier analysis evaluated the prognostic effect of ACTA2-AS1 expression levels. C Forest plot from Cox regression analysis showing patient pathological indicators and ACTA2-AS1 levels as risk factors for an adverse prognosis. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 Low expression of ACTA2-AS1 is associated with an unfavorable prognosis in BCa patients. A Comparison of ACTA2-AS1 expression between BCa patients with favorable and unfavorable prognoses. B Kaplan-Meier analysis evaluated the prognostic effect of ACTA2-AS1 expression levels. C Forest plot from Cox regression analysis showing patient pathological indicators and ACTA2-AS1 levels as risk factors for an adverse prognosis. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 Overexpression of the lncRNA ACTA2-AS1 markedly mitigates malignant biological behaviors in BCa cells To explore ACTA2-AS1’s role in the malignant behavior of BCa cells, we successfully overexpressed it in T24 and UM-UC-3 cells (P < 0.0001, Fig. 3A). Compared to the oe-NC, ACTA2-AS1 overexpression significantly suppressed the proliferation of both T24 and UM-UC-3 cells (P < 0.001, Fig. 3B). Moreover, increased ACTA2-AS1 significantly boosted apoptosis rates (P < 0.001, Fig. 3C). Similarly, it markedly elevated caspase-3 activity (P < 0.001, Fig. 3D). Cell cycle analysis further revealed that, in comparison to oe-NC, cell lines overexpressing ACTA2-AS1 showed a marked decrease in the proportions of cells in S and G2 phases, accompanied by significant G0/G1-phase arrest (P < 0.001, Fig. 3E). Moreover, relative to the oe-NC, overexpressing ACTA2-AS1 notably suppressed the migration and invasion of T24 and UM-UC-3 cells (P < 0.05, Fig. 3F-G). In brief, ACTA2-AS1 overexpression significantly mitigated the malignant biological behavior of BCa cells. Fig. 3Overexpression of the lncRNA ACTA2-AS1 markedly mitigates malignant biological behaviors in BCa cells. A Quantitative assessment of ACTA2-AS1 levels in T24 and UM-UC-3 BCa cells after transfection with an ACTA2-AS1 overexpression plasmid. B CCK-8 assay for evaluating the impact of ACTA2-AS1 overexpression on the proliferation of T24 and UM-UC-3 cells. C-D. Flow cytometry and commercial assay kits for analyzing changes in apoptosis rate and caspase-3 activity. E. Flow cytometry analysis of cell cycle alterations. F-G. Transwell assay to determine the effect of ACTA2-AS1 overexpression on cell migration and invasion. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 Overexpression of the lncRNA ACTA2-AS1 markedly mitigates malignant biological behaviors in BCa cells. A Quantitative assessment of ACTA2-AS1 levels in T24 and UM-UC-3 BCa cells after transfection with an ACTA2-AS1 overexpression plasmid. B CCK-8 assay for evaluating the impact of ACTA2-AS1 overexpression on the proliferation of T24 and UM-UC-3 cells. C-D. Flow cytometry and commercial assay kits for analyzing changes in apoptosis rate and caspase-3 activity. E. Flow cytometry analysis of cell cycle alterations. F-G. Transwell assay to determine the effect of ACTA2-AS1 overexpression on cell migration and invasion. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 ACTA2-AS1 serves as a ceRNA by binding to miR-148b-3p The function of lncRNAs is closely tied to their subcellular localization, with cytoplasmic lncRNAs often acting as ceRNA by targeting miRNAs. To further elucidate the primary role of ACTA2-AS1 in BCa, the lncLocator database predicted its predominant cytoplasmic localization (Fig. 4A). Subcellular fractionation assays subsequently confirmed that ACTA2-AS1 is mainly located in the cytoplasm of T24 and UM-UC-3 cells (P < 0.001, Fig. 4B). Figure 4C illustrates the binding site between ACTA2-AS1 and miR-148b-3p. Transfecting cells with a miR-148b-3p mimic significantly elevated miR-148b-3p levels (P < 0.001, Fig. 4D). Additionally, in DLR assays, miR-148b-3p overexpression notably decreased the luciferase activity of ACTA2-AS1-WT in these two lines, but did not affect that of ACTA2-AS1-MT (Fig. 4E). RIP assay results showed that miR-148b-3p overexpression enhanced the binding of Ago2 to ACTA2-AS1 (P < 0.001, Fig. 4F). Data from Starbase and TCGA databases revealed notably higher miR-148b-3p levels in BCa tumor tissues (P < 0.001, Fig. 4G-H). Among the BCa patients studied, miR-148b-3p levels were significantly higher in tumor tissues than in normal tissues (P < 0.001, Fig. 4I). Moreover, Pearson’s correlation coefficient showed a median negative correlation with ACTA2-AS1 levels in tumor tissues (r = -0.636, P < 0.001, Fig. 4J). Furthermore, miR-148b-3p expression was notably elevated in BCa cell lines compared to SV-HUC-1 cells (P < 0.001, Fig. 4K). Fig. 4ACTA2-AS1 serves as a ceRNA by binding to miR-148b-3p. A The LncLocator database forecasts the subcellular localization of ACTA2-AS1. B Cell fractionation experiments were conducted to determine the subcellular localization of ACTA2-AS1 in T24 and UM-UC-3 cells. C The interaction site between ACTA2-AS1 and miR-148b-3p. D After introducing miR-148b-3p mimic into T24 and MU-UC-3 cells, the miR-148b-3p levels were measured. E-F. DLR and RIP assays validated the specific binding interaction between miR-148b-3p and ACTA2-AS1 (Wild-type [WT] ACTA2-AS1 sequence containing the intact miR-148b-3p binding site [red] and mutant [MT] ACTA2-AS1 sequence with the binding site disrupted [blue] were constructed). G-H. Analysis of miR-148b-3p expression in BCa tumor tissues using Starbase and TCGA databases. I. Among enrolled patients, miR-148b-3p levels were notably different between tumor tissues and adjacent normal tissues. J. Pearson correlation analysis showed a significant inverse relationship between miR-148b-3p and ACTA2-AS1 in tumor tissues. K. miR-148b-3p expression was compared in SV-HUC-1 cells and BCa cell lines. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 ACTA2-AS1 serves as a ceRNA by binding to miR-148b-3p. A The LncLocator database forecasts the subcellular localization of ACTA2-AS1. B Cell fractionation experiments were conducted to determine the subcellular localization of ACTA2-AS1 in T24 and UM-UC-3 cells. C The interaction site between ACTA2-AS1 and miR-148b-3p. D After introducing miR-148b-3p mimic into T24 and MU-UC-3 cells, the miR-148b-3p levels were measured. E-F. DLR and RIP assays validated the specific binding interaction between miR-148b-3p and ACTA2-AS1 (Wild-type [WT] ACTA2-AS1 sequence containing the intact miR-148b-3p binding site [red] and mutant [MT] ACTA2-AS1 sequence with the binding site disrupted [blue] were constructed). G-H. Analysis of miR-148b-3p expression in BCa tumor tissues using Starbase and TCGA databases. I. Among enrolled patients, miR-148b-3p levels were notably different between tumor tissues and adjacent normal tissues. J. Pearson correlation analysis showed a significant inverse relationship between miR-148b-3p and ACTA2-AS1 in tumor tissues. K. miR-148b-3p expression was compared in SV-HUC-1 cells and BCa cell lines. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 LncRNA ACTA2-AS1 acts on miR-148b-3p to reduce the malignancy of BCa cells To further assess the malignant biological impacts of ACTA2-AS1 and miR-148b-3p on BCa cells, a miR-148b-3p mimic was transfected into cells with ACTA2-AS1 overexpression. The reduction in miR-148b-3p levels caused by ACTA2-AS1 overexpression was partly counteracted by the miR-148b-3p mimic (P < 0.0001, Fig. 5A). Additionally, the miR-148b-3p mimic significantly counteracted ACTA2-AS1’s inhibitory effect on T24 and UM-UC-3 cell proliferation (P < 0.001, Fig. 5B). Overexpression of ACTA2-AS1 notably promoted apoptosis in these cells, while the miR-148b-3p mimic weakened this pro-apoptotic effect (P < 0.001, Fig. 5C). Furthermore, miR-148b-3p mitigated the G0/G1 phase cell cycle arrest induced by elevated ACTA2-AS1 in T24 and UM-UC-3 cells, leading to increased proportions of cells in the S and G2 phases (P < 0.001, Fig. 5D). Additionally, while ACTA2-AS1 significantly suppressed cell migration and invasion in these cells, miR-148b-3p markedly weakened this suppressive effect (P < 0.001, Fig. 5E-F). Fig. 5LncRNA ACTA2-AS1 acts on miR-148b-3p to reduce the malignancy of BCa cells. A. Effects of overexpressing ACTA2-AS1 and transfecting with a miR-148b-3p mimic on miR-148b-3p levels in T24 and UM-UC-3 cells. B-C. Utilize CCK-8 assay and flow cytometry to analyze the co-regulatory impacts of ACTA2-AS1 and miR-148b-3p on cell proliferation and apoptosis. D. Analyze cell cycle arrest via flow cytometry. E-F. Employ the Transwell assay to assess the effects of ACTA2-AS1 and miR-148b-3p on cell migration and invasion. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 LncRNA ACTA2-AS1 acts on miR-148b-3p to reduce the malignancy of BCa cells. A. Effects of overexpressing ACTA2-AS1 and transfecting with a miR-148b-3p mimic on miR-148b-3p levels in T24 and UM-UC-3 cells. B-C. Utilize CCK-8 assay and flow cytometry to analyze the co-regulatory impacts of ACTA2-AS1 and miR-148b-3p on cell proliferation and apoptosis. D. Analyze cell cycle arrest via flow cytometry. E-F. Employ the Transwell assay to assess the effects of ACTA2-AS1 and miR-148b-3p on cell migration and invasion. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 DNAJB4 serves as a target gene for miR-148b-3p To identify target genes involved in the ACTA2-AS1/miR-148b-3p interaction, we predicted miR-148b-3p targets using the ENCORI, Tarbase, microT_interactions, Starbase, miRDB, and microT-CDS (Fig. 6A). This analysis revealed overlapping targets, including DNAJB4. Figure 6B shows the binding site between miR-148b-3p and ACTA2-AS1. RIP assay results showed that miR-148b-3p overexpression enhanced the binding of Ago2 to DNAJB4 (P < 0.001, Fig. 6C). Data from Starbase and TCGA databases revealed notably lower DNAJB4 levels in BCa tumor tissues (P < 0.001, Fig. 6D). Moreover, data from GEPIA, Starbase, and TCGA databases indicate a marked reduction in DNAJB4 expression in BCa tumor tissues (Fig. 6E-G). Furthermore, among the recruited BCa patients, DNAJB4 levels in tumor tissues were consistently lower than those in normal tissues (P < 0.001, Fig. 6H). Additionally, Pearson’s correlation coefficient found that DNAJB4 levels in tumor tissue showed a moderate positive correlation with ACTA2-AS1 (r = 0.524) and a moderate negative correlation with miR-148b-3p (r = -0.545, P < 0.001, Fig. 6I-J). Further study demonstrated that ACTA2-AS1 overexpression in T24 and UM-UC-3 cells notably upregulated DNAJB4 expression, but miR-148b-3p significantly weakened this effect (P < 0.001, Fig. 6K). Fig. 6DNAJB4 serves as a target gene for miR-148b-3p. A The Venn diagram displays the overlapping targets of miR-148b-3p, as predicted by ENCORI, Tarbase, microT_interactions, Starbase, miRDB, and microT-CDS databases. B Identification of the binding site between DNAJB4 and miR-148b-3p. C-D. RIP and DLR assays confirmed the targeted interaction between miR-148b-3p and DNAJB4 (Wild-type [WT] DNAJB4 sequence containing the intact miR-148b-3p binding site [red] and mutant [MT] DNAJB4 sequence with the binding site disrupted [blue] were constructed). E-G. GEPIA, Starbase, and TCGA databases were used to predict DNAJB4 expression in BCa tumors and normal tissues. H. Measurement of DNAJB4 expression in tumor tissues from recruited patients. I-J. Correlation analysis of DNAJB4 levels with miR-148b-3p and ACTA2-AS1 in tumor samples. K. Impact of overexpressing ACTA2-AS1 and miR-148b-3p on DNAJB4 levels in T24 and UM-UC-3 cells. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 DNAJB4 serves as a target gene for miR-148b-3p. A The Venn diagram displays the overlapping targets of miR-148b-3p, as predicted by ENCORI, Tarbase, microT_interactions, Starbase, miRDB, and microT-CDS databases. B Identification of the binding site between DNAJB4 and miR-148b-3p. C-D. RIP and DLR assays confirmed the targeted interaction between miR-148b-3p and DNAJB4 (Wild-type [WT] DNAJB4 sequence containing the intact miR-148b-3p binding site [red] and mutant [MT] DNAJB4 sequence with the binding site disrupted [blue] were constructed). E-G. GEPIA, Starbase, and TCGA databases were used to predict DNAJB4 expression in BCa tumors and normal tissues. H. Measurement of DNAJB4 expression in tumor tissues from recruited patients. I-J. Correlation analysis of DNAJB4 levels with miR-148b-3p and ACTA2-AS1 in tumor samples. K. Impact of overexpressing ACTA2-AS1 and miR-148b-3p on DNAJB4 levels in T24 and UM-UC-3 cells. * P < 0.05, ** P < 0.01, *** P < 0.001, ****P < 0.0001 Discussion BCa is the most frequent urinary tract malignancy, with high invasiveness and a notable recurrence rate. Surgical resection is the primary treatment, but chemotherapy and immunotherapy are needed post-surgery to prevent relapse [15]. However, their efficacy is limited. Thus, improving non-invasive prognosis monitoring via molecular biomarkers and predicting patients’ treatment responses are key to advancing precision BCa therapy [16]. Our initial study reveals that ACTA2-AS1 is consistently down-regulated in BCa patients, suggesting its potential as a poor-prognosis biomarker. Moreover, boosting ACTA2-AS1 expression, by targeting the miR-148b-3p/DNAJB4 axis, can impede malignant progression through inhibiting tumor cell proliferation and migration, promoting apoptosis, and triggering cell cycle arrest. Growing evidence reveals that lncRNA is pivotal in modulating BCa malignant progression and acting as a biomarker. For example, lncRNA ELDR is associated with invasive clinicopathological traits in BCa [17]. SNHG20 promotes malignant progression in BCa by affecting cell proliferation and apoptosis [18]. LINC00525 induces invasive phenotypes in BCa [19]. ACTA2-AS1, a newly identified lncRNA, was previously shown to modulate ferroptosis in chondrocytes by suppressing lipid peroxidation [20]. Its role in cancer has also been reported: it inhibits gastric cancer cell growth and promotes apoptosis [21]. In prostate cancer, ACTA2-AS1 expression is notably decreased, and its overexpression significantly inhibits cell proliferation, cell cycle progression, and EMT. Moreover, it is linked to thyroid cancer metastasis [12]. In the early stage, we screened the GEO database to pinpoint differentially expressed lncRNAs in BCa. Surprisingly, ACTA2-AS1 showed consistent downregulation in all four datasets. Even more notably, compared to normal tissues, ACTA2-AS1 was markedly downregulated in BCa tumor tissues across databases like TCGA, GEPIA, and Starbase. Yet, its potential role and clinical mechanisms in BCa are still unknown. In line with prior findings and the research backdrop, our cohort study demonstrated that ACTA2-AS1 is widely downregulated in BCa tumor tissues. Additionally, we found a correlation between its expression levels and worse clinical-pathological indicators in patients. Patients with low ACTA2-AS1 expression seem to have an unfavorable clinical prognosis. Prognosis prediction facilitates stratified patient management and offers insights for customizing treatment plans, ultimately boosting survival rates and elevating the quality of life [22]. Our results suggest that patients with low ACTA2-AS1 expression may require early initiation of treatment or receive more intensive attention, which could potentially optimize their therapeutic outcomes. Most lncRNAs act as molecular sponges in the cytoplasm, competitively binding miRNAs to regulate downstream gene expression [23, 24]. Prior research has demonstrated that miR-148b-3p transferred via exosomes from cancer-associated fibroblasts (CAFs) is notably oncogenic in BCa: it promotes tumor proliferation, metastasis, epithelial-mesenchymal transition (EMT), and chemoresistance, while its downregulation in CAF-derived exosomes enhances BCa cell chemosensitivity [25]. Moreover, Li et al. reported significantly higher miR-148b-3p levels in breast cancer and identified a miRNA signature trio, including miR-148b-3p, suitable for BC diagnosis [26]. Additionally, miR-148b-3p was also notably upregulated in tumor samples from 158 BCa patients [27]. In line with prior research, we detected a significant rise in miR-148b-3p expression in BCa tissues, consistent with bioinformatics database predictions. Notably, this study is the first to show that elevated miR-148b-3p counteracts the effects of ACTA2-AS1 in reducing BCa cell proliferation, migration, and invasion, while also inhibiting apoptosis and cell cycle arrest. ACTA2-AS1 might impede malignant progression in BCa by targeting miR-148b-3p to weaken its malignant biological activities. DNAJB4, a member of the DNAJ protein family, serves as a co-chaperone for HSP70 proteins and has been investigated in various cancers. For example, it is markedly downregulated in triple-negative breast cancer and is associated with a poor prognosis [28]. In non-small cell lung cancer, downregulated DNAJB4 promotes malignant biological behavior [29]. Moreover, low DNAJB4 expression in gastric cancer correlates with a poor prognosis and induces apoptosis through caspase-3 [30]. In renal cell carcinoma, DNAJB4 shows widespread reduction and is linked to the tumor immune microenvironment and drug sensitivity [31]. DNAJB4 also exerts a significant influence on BCa. Ke et al. (2021) retrieved differentially expressed mRNAs in BCa from the GSE13507 dataset and developed a seven -gene panel, including DNAJB4. This panel can predict adverse outcomes in BCa and is strongly linked to immune cell infiltration [32]. Moreover, Guang et al. (2025) also identified DNAJB4 as a macrophage-related gene, which, in combination with six additional genes, could potentially enhance the metastatic potential of bladder cancer [33]. Moreover, research has shown that miR-4454, linked to human endogenous retroviruses, drives BCa progression by suppressing DNAJB4 expression [34]. In our study, data from GEPIA, StraBase, and TCGA confirmed that DNAJB4 is notably downregulated in BCa tumor tissues. Aligning with these prior findings, we also observed a significant decrease in DNAJB4 levels within BCa carcinomas. Immortally, our study is the first to suggest that DNAJB4 may act as a target gene regulated by the ACTA2-AS1/miR-148b-3p axis, contributing to BCa’s malignant progression. Compared with previous studies, this work first identified the ACTA2-AS1/miR-148b-3p/DNAJB4 regulatory axis in BCa, which enriches the ceRNA regulatory network and provides a novel research basis for bladder cancer. However, these findings require further validation. It cannot be denied that this study has certain limitations. First, the optimal cutoff value for ACTA2-AS1 expression was determined based on the mean level in our single-center cohort, which may not be directly applicable to other populations or larger patient groups. Future large-scale, multicenter validation studies are therefore required to confirm this cutoff value and verify the generalizability of our findings. Moreover, this study only performed cellular experiments in vitro, lacking animal experimental verification and validation in large independent clinical cohorts. Additionally, beyond the mechanistic role in BCa progression, ACTA2-AS1 also presents certain clinical translation value as a prognostic biomarker. However, several practical limitations remain for its clinical application. LncRNA molecules generally have poor RNA stability, which may affect the accuracy and reproducibility of detection in clinical specimens. Moreover, unified technical specifications and detection standardization are still lacking. Further multicenter validation and standardized testing systems are needed to improve the clinical practicability of ACTA2-AS1. In brief, the present study suggests that decreased ACTA2-AS1 expression may serve as a potential prognostic biomarker for unfavorable outcomes in patients with BCa. Additionally, it regulates BCa cell behavior through the miR-148b-3p/DNAJB4 pathway, thereby curbing malignant progression. Our findings may provide new therapeutic strategies for BCa treatment.