Cancer Information Group

Menu

(Article In Press) The mechanism of HNRNPF enhancing the stability of DUSP5 mRNA to inhibit cellular senescence in bladder cancer

By /

Qixin Mo1, Yu Wang1, Henghui Zhang1, Xinlei Zhao1, Hang Su1, Dongqing Li1, Chen Chen1, Fei Li1, *

1Department of Urology, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, People’s Republic of China

*Corresponding author: Fei Li (feili20700338@126.com)

Present address: No. 1838 North Guangzhou Avenue, Baiyun District, Guangzhou, People’s Republic of China

 

Abstract

Background: Bladder cancer (BCa) is a prevalent malignancy characterized by poor prognosis. The incidence of BCa is gradually rising while the five-year overall survival (OS) rate remains unexpected, especially in patients with muscle-invasive BCa.Cellular senescence serves as a potent physiological barrier against tumorigenesis, yet the mechanisms by which BCa cells evade this process remain elusive. Heterogeneous Nuclear Ribonucleoprotein F (HNRNPF), as a classical RNA-binding protein, has been implicated in RNA processing and tumor progression, but its specific role in modulating senescence in BCa require further elucidation.

Methods: Four bladder cancer cell lines and clinical tissue specimens were utilized in this study. We performed quantitative real-time PCR (qRT-PCR), Western blotting, immunohistochemistry (IHC), SA-β-Gal staining, Methylated single-stranded RNA affinity assays, animal xenograft model studies, bioinformatic analyses, and RNA stability assays.

Results: HNRNPF inhibited BCa cellular senescence and promoted tumor progression both in vitro and in vivo. Through bioinformatic prediction and RNA-sequencing screening, dual-specificity phosphatase 5 (DUSP5) was identified and verified as a downstream target of HNRNPF. Further investigation revealed that DUSP5 mediates HNRNPF-regulated cellular senescence in BCa. Mechanistically, the HNRNPF protein binds directly to DUSP5 mRNA and may regulate DUSP5 in an m6A-dependent manner.

Conclusions: HNRNPF suppresses bladder cancer cellular senescence by enhancing DUSP5 mRNA stability, potentially via an m6A-dependent mechanism.

Keywords: Bladder cancer; Cellular senescence; HNRNPF; m6A; Stability

Introduction

Bladder cancer (BCa) is one of the most prevalent malignancies of the urinary system[1, 2]. As indicated by global cancer statistics, the incidence of the disease is rising, with the number of diagnosed cases expected to double by 2040[3, 4]. Clinically, urothelial carcinoma is primarily classified into two distinct categories: non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC)[5]. Notwithstanding the continuous advancements witnessed in radical surgical approaches and pharmacological treatment regimens in recent years, the five-year overall survival (OS) rate for BCa patients remains unsatisfactory, particularly in patients with MIBC. Moreover, there is an absence of specific biomarkers with independent prognostic value[6]. Consequently, elucidating the molecular mechanisms underlying BCa initiation and progression is of significant importance for improving patient survival and clinical outcomes.

Cellular senescence is defined as a stable, terminal state of cell cycle arrest that physiologically functions as a cellular stress response aimed at eliminating damaged cells to maintain tissue homeostasis[7, 8]. It is posited that senescence functions as a defense mechanism against cancer. In the event of encountering stress stimuli, such as persistent DNA damage, tumour cells undergo a process known as senescence. This is a physiological response intended to halt further genomic instability and the accumulation of DNA damage[9, 10]. Consequently, senescence functions as an anti-tumour barrier. Paradoxically, cellular senescence has also been reported to create a pro-tumorigenic microenvironment and drive tumour development through the senescence-associated secretory phenotype (SASP)[11]. Presently, therapeutic targeting of cellular senescence has been employed, in a preliminary capacity, as one of the clinical anti-tumour strategies. When employed in conjunction with specific pharmacological agents as part of tumour treatment regimens, this approach has the potential to enhance drug sensitivity while minimizing drug toxicity[12].

The initiation and progression of BCa involve complex gene regulatory networks, in which RNA-binding proteins (RBPs) and RNA modifications play central roles in post-transcriptional regulation[13]. It is important to note that this post-transcriptional regulatory network is subject to further refinement by RNA modifications. Among these modifications, N6-methyladenosine (m6A) represents the most abundant internal mRNA modification in mammals. m6A exerts a significant influence on a variety of processes, including RNA splicing, stability, and translation efficiency[14]. The dynamic balance of this modification is maintained by methyltransferases (writers), demethylases (erasers), and reader proteins. Substantial evidence has demonstrated that m6A modification-associated proteins, such as METTL3, METTL14, and ALKBH5, are deeply involved in BCa proliferation, metastasis, and chemotherapy resistance through the regulation of mRNA metabolism of key oncogenes or tumour suppressor genes[15, 16]. This finding indicates that m6A reader proteins and their role in regulating RNA metabolism may be a significant factor in understanding the pathological mechanisms of BCa.

Heterogeneous nuclear ribonucleoprotein F (HNRNPF) belongs to the hnRNP family, a class of RNA-binding proteins (RBPs)[17]. Multiple family members have been reported to function as reader proteins[18, 19]. It has been established through previous studies that HNRNPF has a role in promoting the growth of BCa cells, as well as their ability to invade and metastasize. This process is facilitated by the control of signaling axes, including TPX2/FOXO1 and Snail1[19-21]. However, the role of HNRNPF in BCa cellular senescence and its potential mechanism as an m6A reader protein have not been fully elucidated, and other functions, roles, and mechanisms of HNRNPF remain to be explored. In 2018, Huang et al. reported that HNRNPF binds to artificial m6A consensus sequences with high affinity, providing a novel research perspective[22]. In consideration of the findings outlined above, the present study seeks to ascertain whether HNRNPF exerts a regulatory influence on the mRNA stability of its downstream target gene DUSP5 via an m6A-dependent mechanism, thereby impeding the progression of BCa cellular senescence. The findings of this study will contribute to the revelation of novel mechanisms of HNRNPF in bladder cancer and provide a theoretical basis for senescence-based therapeutic strategies for BCa.

Materials and Methods

Collection of clinical specimens

Between April 2020 and August 2023, paired samples of tumor and matched adjacent normal bladder mucosa were collected from 14 patients undergoing radical cystectomy for urothelial carcinoma at the Department of Urology, Nanfang Hospital, Southern Medical University. All experimental protocols involving human subjects were reviewed and approved by the Ethics Committee of Nanfang Hospital. The study strictly adhered to the principles of the Declaration of Helsinki, and written informed consent was obtained from every participant prior to sample acquisition.

Cell culture  

The immortalized human urothelial cell line SV-HUC-1 and a panel of bladder cancer (BCa) cell lines—specifically T24, RT4, SW780, and UMUC-3—were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Prior to experimentation, all cell lines underwent authentication to verify their identity. Maintenance of the BCa lines was performed in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco), whereas SV-HUC-1 cells were propagated in RPMI-1640 medium (Gibco). Both media formulations were supplemented with 10% fetal bovine serum (FBS) and a 1% penicillin/streptomycin antibiotic mixture. Cultures were kept in a humidified atmosphere containing 5% CO2 at 37 °C.

Bioinformatics analysis

Transcriptomic profiles of bladder cancer (BCa) patients were retrieved from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) repositories. We categorized samples based on muscle invasion status (NMIBC vs. MIBC) and subsequently stratified them into high- and low-HNRNPF expression cohorts. Differentially expressed genes (DEGs) were identified using stringent thresholds of a false discovery rate (FDR) 2. To predict direct downstream targets, we utilized POSTAR2 (http://lulab.life.tsinghua.edu.cn/postar), a robust platform for analyzing RNA-binding protein interactions. Finally, Venn diagram analysis was performed to isolate overlapping genes shared between the POSTAR2 binding targets and the DEGs identified in the clinical datasets.

Protein extraction and Western blot

Total protein was isolated from both tissue specimens and cell cultures using RIPA lysis buffer containing a cocktail of protease and phosphatase inhibitors (Fude Biological, Hangzhou, China). Lysates were mixed with loading buffer and boiled at 100 °C for 5 minutes to ensure denaturation. Proteins were then resolved via SDS-PAGE and electro-transferred to polyvinylidene fluoride (PVDF) membranes (Millipore). To prevent non-specific binding, membranes were incubated in a 5% rapid blocking solution (Fude Biological) for 20 minutes before being probed with specific primary antibodies at 4 °C overnight. This was followed by a 2-hour incubation with the corresponding secondary antibodies (ProteinTech). β-actin and GAPDH served as the internal loading control. Images were captured using an automated imaging system, ensuring all comparative data were derived from parallel experiments under identical conditions. The Western blot images were captured using the Tanon-4600 imaging system (Biotanon). The following primary antibodies were used: anti-HNRNPF (67701-1-Ig, ProteinTech), anti-DUSP5(K008035P, Solarbio), anti-P53 (10442-1-AP, ProteinTech), anti-P-Rb (T59027, Abmart), anti-Rb (TB3384, Abmart), anti-P21 (82669-2-RR, ProteinTech), anti-yH2AX (83307-2-RR, ProteinTech), anti-CDK2(10122-1-AP, ProteinTech) and anti-CDK4(11026-1-AP, ProteinTech).

Transfection and establishment of stable clone cells, cell transfection

To generate stable models of HNRNPF modulation, lentiviral particles carrying either HNRNPF overexpression (oe-HNRNPF) or shRNA constructs (sh-HNRNPF) were transduced into specific bladder cancer cell lines. T24 and SW780 cells were selected for overexpression, yielding the T24-oeHNRNPF and SW780-oeHNRNPF sublines, respectively. Conversely, knockdown lines (UMUC-3-shHNRNPF and RT4-shHNRNPF) were established from parental UMUC-3 and RT4 cells. Corresponding empty vector controls (oe-NC and sh-NC) were simultaneously generated for all conditions. Where transient manipulation was required, plasmid transfections were performed utilizing the Lipofectamine 3000 reagent (Invitrogen, USA) according to the manufacturer’s protocol.

Subcutaneous xenograft model

All animal procedures were reviewed and approved by the Institutional Animal Ethics Committee of Nanfang Hospital, Southern Medical University, and adhered strictly to national welfare guidelines. Four-to-five-week-old male BALB/c nude mice were housed under specific pathogen-free (SPF) conditions. For the xenograft model, mice were randomly allocated into four cohorts. Log-phase BCa cells were harvested to prepare single-cell suspensions at a density of 1 × 107/ml. Subcutaneous injections were performed as follows: T24-oeNC (control) vs. T24-oeHNRNPF, and T24-shNC (control) vs. T24-shHNRNPF. Throughout the study, tumor progression was monitored by measuring dimensions and body weight at regular intervals. The endpoint criterion for tumor burden was strictly observed according to ethical mandates. Four weeks post-inoculation, mice were anesthetized and sacrificed; tumors were excised, weighed, and photographed for subsequent analysis.

Real-time PCR 

Total RNA isolation was performed from both patient tissues and cultured cells using the TRIzol reagent in strict accordance with the manufacturer’s protocol. Subsequently, 1 µg of the isolated RNA was reverse-transcribed into cDNA using the High-Capacity RNA-to-cDNA Kit (Applied Biosystems). Gene expression levels were quantified via real-time PCR (qPCR) on an Applied Biosystems 7500 Fast System. All reactions were carried out using the SYBR Green Master Mix and specific primers, following standard thermal cycling conditions.

IHC staining 

Following standard protocols, paraffin-embedded tumor and tissue samples underwent deparaffinization and rehydration. After antigen retrieval and a blocking step, the specimens were probed overnight at 4 °C with primary antibodies against HNRNPF (67701-1-Ig, ProteinTech), DUSP5 (K008035P, Solarbio), or P53 (10442-1-AP, ProteinTech). This was followed by incubation with corresponding secondary antibodies. Finally, signal detection was performed using a DAB substrate kit (Thermo Fisher Scientific) to visualize immunoreactivity.

Cell counting kit-8 assay

To evaluate cell viability, we utilized the Cell Counting Kit-8 (CCK-8) assay in accordance with the manufacturer’s guidelines (Dojindo). Cells were plated into 96-well plates at an initial density of 1,000 cells per well. Following transfection, the cells were cultured for varying durations: 0, 24, 48, 72, and 96 hours. At each time point, 10 μl of CCK-8 reagent was introduced to each well, followed by a 1.5-hour incubation at 37 °C. Optical density was subsequently determined by measuring absorbance at 450 nm with a spectrophotometer (Thermo Fisher Scientific).

SA-β-Gal Staining

To assess SA-β-gal activity, we processed both 5 µm frozen tumor sections and cells grown on coverslips according to the manufacturer’s protocol (KGA5101-100, KeyGEN BioTECH). The tissue sections underwent additional counterstaining with hematoxylin. Following image acquisition, we used ImageJ software to quantify the proportion of SA-β-gal-positive cells or the positive area ratio.

RNA immunoprecipitation (RIP)

We conducted RNA immunoprecipitation (RIP) assays utilizing the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Billerica, MA, USA) in line with the manufacturer’s directions. In brief, harvested cells were pelleted and resuspended in 200 µl of lysis buffer supplemented with a protease inhibitor cocktail (1 µl) and an RNase inhibitor (0.5 µl). Magnetic beads were first conjugated with either anti-rabbit IgG or anti-rabbit HNRNPF antibodies for one hour at room temperature. The cell lysates were then incubated with these bead-antibody complexes overnight at 4 °C. Finally, RNA was extracted from the immunoprecipitated RNA-protein complexes and subsequently analyzed via RT-qPCR and agarose gel electrophoresis.

Methylated single-stranded RNA affinity assay

To investigate m6A-binding interactions, we utilized synthesized single-stranded RNA oligonucleotide probes featuring the GGACU consensus sequence. These probes, containing either methylated adenosine (ss-m6A: 5’-biotin-CGUCUCGG(m6A) CUCGG(m6A) CUGCU-3’) or unmethylated adenosine (ss-A: 5’-biotin-CGUCUCGGACUC GGACUGCU-3’), were produced and verified via mass spectrometry by Tsingke Biotech Company (Beijing, China). Following the standard protocol of an RNA pull-down kit (Thermo Fisher Scientific), the biotinylated probes were anchored to streptavidin magnetic beads and incubated with total protein lysates from UMUC-3 cells for 8 hours at 4 °C. After washing the beads twice to remove non-specific binding, the probe-bound proteins were eluted, resolved on 10% SDS-PAGE gels, and subsequently visualized using silver staining and western blotting.

Silver staining of protein gels

Following gel electrophoresis, protein bands were visualized using the Protein Fast Silver Stain Kit (P0017S, Beyotime) according to the provided instructions. The intensity of the silver-stained bands was subsequently captured and analyzed using a Bio-Rad scanner.

Measurement of mRNA stability

To determine mRNA stability, stable cell lines (T24-oeNC/oeHNRNPF and RT4-shNC/shHNRNPF) were treated with actinomycin D (5 μg/mL; APExBIO, TX, USA) to inhibit transcription. Cells were harvested at designated post-treatment time points (0, 4, 8, 12, 18, and 24 hours), followed by RNA isolation. The half-life of DUSP5 mRNA was then quantified via qRT-PCR, using the protocols previously outlined.

Statistical analysis

Data are expressed as the mean ± SD based on a minimum of three independent replicates. Statistical comparisons between two groups were conducted using unpaired, two-tailed Student’s t-tests with GraphPad Prism 8 software; a P-value of less than 0.05 was deemed statistically significant. The Kaplan–Meier method was utilized for survival analysis. Significance levels are denoted as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, with n.s. representing non-significant results.

Results

HNRNPF inhibits Cellular Senescence by BCA cells in vitro and in vivo

Bioinformatics analysis of TCGA data stratified BCa patients into HNRNPF-high and -low expression groups. KEGG pathway analysis revealed enrichment of DNA repair and replication pathways in the high-expression cohort (Fig. 1A). GSEA further identified a marked correlation between HNRNPF expression and DNA damage-induced cellular senescence signatures (Fig. 1B), suggesting a regulatory role in genomic stability. To validate these findings, HNRNPF was stably overexpressed in T24/SW780 cells and silenced in UMUC-3/RT4 cells. qPCR and Western Blot confirmed significant upregulation or knockdown efficiency compared to controls (Fig. 1C, D). Phenotypic assessment via Senescence-Associated β-Galactosidase (SA-β-gal) staining demonstrated that HNRNPF overexpression significantly reduced the percentage of SA-β-gal-positive cells. Conversely, HNRNPF silencing induced a senescent morphology with widespread SA-β-gal staining (Fig. 1E), indicating that HNRNPF is essential for averting senescence. Mechanistically, Western Blot analysis showed that HNRNPF upregulation suppressed the CDK inhibitor p21 while maintaining CDK2/4 levels. In contrast, HNRNPF-silenced cells exhibited elevated p21, confirming cell cycle arrest (Fig. 1F). Consistent with in vitro data, xenograft analysis revealed that HNRNPF overexpression repressed, whereas depletion promoted, cellular senescence in vivo (Fig. 1G, H). These results collectively imply that HNRNPF exerts its oncogenic function by inhibiting the cellular senescence program in BCa.

Figure 1 HNRNPF suppresses cellular senescence in vitro and in vivo.

(A) KEGG pathway enrichment analysis of differentially expressed genes (DEGs) between HNRNPF-high and HNRNPF-low groups in the TCGA-BLCA cohort, highlighting enrichment in DNA replication and repair pathways. (B) Gene Set Enrichment Analysis (GSEA) plot showing a positive correlation between high HNRNPF expression and the DNA damage-induced cellular senescence gene signature. (C) Validation of HNRNPF overexpression and knockdown efficiency at the mRNA level in T24/SW780 (overexpression) and RT4/UMUC-3 (knockdown) cells using qPCR. Data are presented as mean ± SD; *p < 0.05 vs. Control. (D) Western blot verification of HNRNPF protein levels in the constructed stable cell lines. (E) Representative images and quantitative analysis of Senescence-associated β -galactosidase (SA-β-gal) staining in BCa cells. HNRNPF overexpression reduced, while knockdown increased, the percentage of SA-β-gal -positive cells. (F) Western blot analysis of cell cycle and senescence-related proteins (p21, CDK2, CDK4) in HNRNPF-modulated cells, showing that HNRNPF knockdown induces p21 accumulation. (G-H) SA-β-gal staining of frozen tumor sections from xenografts. HNRNPF silencing significantly increased senescence in vivo, which is statistically verified.

Identification of HNRNPF targets in BCA

To elucidate the mechanism by which HNRNPF inhibits cellular senescence, we integrated high-throughput sequencing data to screen for potential downstream targets. Intersection analysis was performed using three datasets: HNRNPF-binding targets from the POSTAR2 database (2,507 mRNAs), differentially expressed genes (DEGs) between BCa and normal tissues from TCGA (Fig. 2A), and RNA-seq data from HNRNPF-knockdown cells (Additional file 1: Table S1). Venn diagram analysis identified 20 overlapping candidate genes shared across all profiles. Among these, Dual Specificity Phosphatase 5 (DUSP5) was selected for further validation due to its significant enrichment in relevant pathways.

Validation in clinical samples confirmed a strong correlation. qPCR and immunohistochemistry (IHC) quantification of fresh BCa tissues revealed that HNRNPF and DUSP5 expression levels were significantly positively correlated (Fig. 2B, C). High HNRNPF expression consistently mirrored elevated DUSP5 levels in patient samples.          To confirm this regulatory relationship, we assessed DUSP5 expression in our modulated cell lines. Western Blot and qPCR analyses demonstrated that DUSP5 was significantly upregulated in HNRNPF-overexpressing cells but downregulated in HNRNPF-silenced cells in vitro (Fig. 2D, E) and in xenograft tissues in vivo (Fig. 2F). Furthermore, HNRNPF expression strongly correlated with p53 and yH2AX, established biomarkers of DNA damage repair and senescence, suggesting a potential mechanistic link. Collectively, these results indicate that HNRNPF functions as a positive regulator of DUSP5 in BCa, implying that DUSP5 is a critical downstream effector mediating the oncogenic effects of HNRNPF.

Figure 2: HNRNPF positively regulates DUSP5 expression in BCa tissues and cell lines.

(A) Venn diagram illustrating the identification of potential HNRNPF downstream targets by intersecting POSTAR2 binding data, TCGA differentially expressed genes, and RNA-seq data from HNRNPF-knockdown cells. DUSP5 was identified as a key candidate. (B) Correlation analysis of HNRNPF and DUSP5 mRNA expression levels in fresh clinical BCa tissues, assessed by qPCR. (C) Representative Immunohistochemistry (IHC) staining images and statistical quantification of HNRNPF and DUSP5 protein expression in clinical BCa specimens (n=15), showing a positive correlation. Scale bar: 50 μm. (D) Western blot analysis determining DUSP5 protein levels in HNRNPF-overexpressing (T24, SW780) and HNRNPF-knockdown (RT4, UMUC-3) cells. (E) qPCR analysis of DUSP5 mRNA levels in the indicated BCa cell lines. (F) Western blot analysis of HNRNPF, DUSP5, and senescence markers (p53, yH2AX) in key tumor tissues derived from the xenograft mouse model.

DUSP5 Suppresses Cellular Senescence and Promotes Proliferation in BCA

To investigate the specific role of DUSP5 in bladder cancer progression, we testified the expression of DUSP5 in BCa cell lines (Fig. 3A, B) and transiently silenced DUSP5 using small interfering RNAs (si-DUSP5-1 and si-DUSP5-2) in T24 and RT4 cells, where DUSP5 highly expressed. The knockdown efficiency was validated at both transcriptional and translational levels, showing significantly reduced DUSP5 expression compared to the negative control (si-NC) (Fig. 3C, D). Phenotypic assays revealed that DUSP5 functions as a critical inhibitor of senescence. Senescence-associated SA-β-gal staining demonstrated that transient silencing of RT4 cell line resulted in a substantial increase in SA-β-gal-positive cells (Fig. 3E). Furthermore, Cell Counting Kit-8 (CCK-8) assays showed that DUSP5 knockdown significantly impaired cell viability and proliferation rates in BCa cells compared to controls (Fig. 3F, G). These findings suggest that endogenous DUSP5 is essential for maintaining the proliferative capacity of bladder cancer cells by averting premature senescence. Collectively, these data confirm that DUSP5 acts as an oncogene in BCa, promoting tumor progression by suppressing cellular senescence.

DUSP5 deficiency restores the cellular senescence inhibition effects exerted by HNRNPF overexpression in BCA cells

To determine whether DUSP5 is the functional mediator of HNRNPF-induced cellular senescence, rescue experiments were performed. We transiently silenced DUSP5 in BCa cells stably overexpressing HNRNPF. The efficiency of DUSP5 knockdown in the HNRNPF-OE background was confirmed by qPCR and Western blotting (Fig. 3H, K) prior to functional assays. Phenotypically, while HNRNPF overexpression significantly promoted cell proliferation and suppressed cellular senescence in T24 and SW780 cells, concurrent knockdown of DUSP5 effectively reversed these effects. Specifically, DUSP5 depletion abrogated the growth advantage conferred by HNRNPF (Fig. 3I) and restored the senescent phenotype, as evidenced by increased SA-β-gal staining (Fig. 3J). At the molecular level, Western blot analysis revealed that DUSP5 knockdown counteracted HNRNPF-induced alterations in cell cycle-related proteins, notably restoring p21 expression (Fig. 3K). These findings confirm that DUSP5 is a critical downstream effector of HNRNPF, and the HNRNPF/DUSP5 axis facilitates bladder cancer progression by inhibiting the senescence program.

Figure 3: DUSP5 is a functional downstream target of HNRNPF mediating cellular senescence.

(A) Quantitative Real-Time PCR (qPCR) analysis of DUSP5 mRNA expression levels across a panel of bladder cancer cell lines. (B) Western blot analysis determining endogenous DUSP5 protein levels in various BCa cell lines. Based on these profiles, T24 and RT4 cells were selected for subsequent knockdown experiments. (C) Validation of DUSP5 knockdown efficiency at the transcriptional level. Cells were transiently transfected with si-DUSP5-1, si-DUSP5-2, or a negative control (si-NC), and mRNA levels were assessed by qPCR. (D) Western blot confirmation of DUSP5 protein depletion in the transfected BCa cells. (E) Representative images and quantitative analysis of Senescence-Associated β-Galactosidase (SA-β-gal) staining. Transient silencing of DUSP5 resulted in a significant increase in the proportion of SA-β-gal -positive (senescent) cells. Scale bar: 100 μm. (F-G) Cell proliferation curves assessed by Cell Counting Kit-8 (CCK-8) assay. DUSP5 knockdown significantly impaired the viability and growth rate of BCa cells compared to the control group over 96 hours. Data are presented as mean ± SD; *p < 0.05. (H) qPCR validation of DUSP5 knockdown efficiency in T24 and SW780 cells stably overexpressing HNRNPF. (I) CCK-8 proliferation assay showing that silencing DUSP5 reverses the growth-promoting effect of HNRNPF overexpression. (J) Representative SA-β-gal staining images indicate that DUSP5 knockdown restores the senescence phenotype in HNRNPF-overexpressing cells. (K) Western blot analysis showing that DUSP5 depletion reverses the HNRNPF-mediated suppression of p21, confirming that the HNRNPF-DUSP5 axis regulates the cell cycle machinery to mediate cellular senescence.

HNRNPF acts as an m6A-reader protein and may stabilize DUSP5 mRNA in an m6A-related manner in BCA

To elucidate the precise molecular mechanism by which HNRNPF regulates DUSP5, we first examined their physical interaction. RNA Immunoprecipitation (RIP) assays using an anti-HNRNPF antibody confirmed that HNRNPF protein directly binds to endogenous DUSP5 mRNA (Fig. 4A). Following this, we assessed mRNA stability in HNRNPF-modulated cells treated with Actinomycin D. Our results demonstrated that the half-life of DUSP5 mRNA was significantly prolonged in HNRNPF-overexpressing cells (Fig. 4B), whereas it was markedly shortened in HNRNPF-knockdown cells (Fig. 4C). These data suggest that HNRNPF positively regulates DUSP5 expression by enhancing its mRNA stability. Given that HRNPNF is an RNA-binding protein, we hypothesized an m6A-dependent mechanism. Bioinformatics analysis predicted multiple high-confidence N6-methyladenosine (m6A) sites within the DUSP5 mRNA sequence (Fig. 4D). Consistent with this, TCGA database analysis revealed a significant positive correlation between DUSP5 expression and key m6A writers, including METTL3 and RBMX, in BCa tissues (Fig. 4E). To validate this interaction on DUSP5 specifically, we performed methylated RNA immunoprecipitation (meRIP)-qPCR. The results showed significant enrichment of DUSP5 mRNA using m6A-specific antibodies compared to IgG controls, confirming that DUSP5 is indeed m6A-modified (Fig. 4F). To determine if HNRNPF functions as an m6A reader, we performed RNA pull-down assays using biotinylated single-stranded RNA probes containing the consensus m6A sequence (ss-m6A) or an unmethylated control (ss-A). Immunoblotting and silver staining revealed that HNRNPF bound selectively and with higher affinity to the methylated probe (Fig. 4G), identifying it as a specific m6A-binding protein. Taken together, these findings indicate that HNRNPF may act as an m6A reader that recognizes m6A sites on DUSP5 transcripts to enhance their stability.

Figure 4: HNRNPF acts as an m6A reader to stabilize DUSP5 mRNA.

(A) RNA Immunoprecipitation (RIP) assay performed with anti-HNRNPF antibody followed by qPCR, demonstrating the direct binding of HNRNPF protein to DUSP5 mRNA. IgG was used as a negative control. (B–C) mRNA stability assays in BCa cells treated with Actinomycin D. DUSP5 mRNA half-life was prolonged in HNRNPF-overexpressing cells (B) and shortened in HNRNPF-knockdown cells (C). (D) Schematic representation of predicted N6-methyladenosine (m6A) sites on the DUSP5 mRNA sequence. (E) Correlation analysis from the TCGA database showing a significant positive relationship between DUSP5 expression and key m6A writers (METTL3, RBMX). (F) Methylated RNA Immunoprecipitation (meRIP)-qPCR analysis confirming m6A modification abundance on DUSP5 mRNA transcripts. (G) RNA pull-down assay followed by Western blotting and silver staining. HNRNPF protein was pulled down by the methylated single-stranded RNA probe (ss-m6A) but not by the unmethylated control (ss-A).

Discussion

Bladder cancer (BCa) remains a significant global health burden, characterized by high recurrence rates and limited therapeutic options for advanced disease[23]. Despite advances in multimodal treatments, the 5-year survival rate for muscle-invasive bladder cancer has plateaued, underscoring the urgent need to decipher the molecular drivers of BCa progression. In this study, we identified Heterogeneous Nuclear Ribonucleoprotein F (HNRNPF) as an oncogenic driver that facilitates tumor growth by averting cellular senescence. Mechanistically, we demonstrated for the first time that HNRNPF serves as an m6A reader, stabilizing DUSP5 mRNA to inhibit the p53/p21-mediated senescence pathway. These findings not only unravel a new layer of post-transcriptional regulation in BCa but also highlight the HNRNPF-DUSP5 axis as a potential therapeutic target.

Cellular senescence acts as a potent physiological barrier against tumorigenesis, permanently halting the proliferation of cells at risk of neoplastic transformation[24]. Overcoming this barrier is a prerequisite for cancer development. Our study reveals that HNRNPF overexpression allows BCa cells to bypass this checkpoint. We observed that silencing HNRNPF triggered a classic senescence phenotype, characterized by cell cycle arrest, enlarged morphology, and increased SA-β-gal activity both in vitro and in vivo. This suggests that high levels of HNRNPF shield tumor cells from endogenous stress signals that would otherwise induce growth arrest. By suppressing senescence, HNRNPF ensures the maintenance of a high proliferative index, a hallmark of aggressive bladder carcinomas. This aligns with recent “pro-senescence” therapeutic strategies, which aim to reactivate this dormant program to halt tumor progression.

A key discovery of our work is the identification of Dual Specificity Phosphatase 5 (DUSP5) as a critical downstream effector of HNRNPF. DUSP5 acts as a negative regulator of the MAPK/ERK pathway, a signaling cascade often hyperactivated in cancer to drive proliferation[25, 26]. Interestingly, while some reports suggest DUSP5 can be a tumor suppressor, our data indicates that in the specific context of HNRNPF-driven BCa, DUSP5 is upregulated and essential for senescence evasion. This context-dependence is common in signaling networks. Our rescue experiments confirmed that DUSP5 knockdown could reverse the phenotype caused by HNRNPF overexpression, restoring p21 level. This implies that the oncogenic function of HNRNPF is heavily reliant on maintaining high levels of DUSP5, possibly to fine-tune ERK signaling to a level that supports proliferation without triggering oncogene-induced senescence[27].

The most novel aspect of our study lies in the mechanistic elucidation of how HNRNPF regulates DUSP5. We identified HNRNPF as a specific reader of N6-methyladenosine (m6A) modifications. m6A is the most prevalent internal modification in eukaryotic mRNA and plays a pivotal role in regulating RNA fate, including stability, splicing, and translation[28]. While METTL3 and other “writers” have been extensively studied in BCa, the role of specific “readers” remains less defined. Our metabolic labeling and immunoprecipitation assays conclusively showed that HNRNPF binds to m6A sites on DUSP5 mRNA, thereby extending its half-life. This adds HNRNPF to the growing list of functional m6A readers and provides a precise molecular explanation for the concordant expression of HNRNPF and DUSP5 observed in clinical samples.

The clinical implications of these findings are substantial. Since HNRNPF is significantly upregulated in tumor tissues compared to normal urothelium, it holds promise as a diagnostic biomarker. Furthermore, the HNRNPF-m6A-DUSP5 axis represents a vulnerability that could be exploited therapeutically. Targeting the RNA-binding domain of HNRNPF or interfering with its interaction with m6A sites could destabilize DUSP5 and force tumor cells into senescence. Unlike apoptosis-inducing chemotherapy, pro-senescence therapy might offer an alternative approach with lower toxicity, potentially sensitizing tumors to subsequent immune clearance via the senescence-associated secretory phenotype (SASP)[29, 30].

Despite these promising findings, our study has limitations that must be acknowledged. First, while we validated the interaction between HNRNPF and DUSP5 mRNA using RIP and meRIP assays, we did not perform site-directed mutagenesis of the specific m6A residues within the DUSP5 transcript within limited time. Constructing DUSP5 mutants with abolished m6A sites would provide definitive evidence that HNRNPF-mediated stabilization relies strictly on these specific methylation events. Second, given the complexity of RNA regulation, it is possible that HNRNPF modulates the stability or splicing of other senescence-related transcripts beyond DUSP5, which warrants high-throughput screening in future studies. Finally, the specific signaling events downstream of DUSP5 that directly mediates p21 and other cell cycle protein in this context need further detailed mapping.

In conclusion, our study establishes HNRNPF as a critical regulator of bladder cancer progression and resistance to senescence. By elucidating the mechanism wherein HNRNPF acts as an m6A reader to stabilize DUSP5 mRNA, we connect epitranscriptomic regulation with cell cycle control. This work not only deepens our understanding of BCa biology but also suggests that disrupting the HNRNPF/DUSP5 axis could serve as a novel strategy to induce senescence and halt tumor growth.

Acknowledgements

We thank all individuals participated in this study.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

All authors contributed to the work presented in this paper. Qixin Mo: Conceptualization, investigation, visualization, and writing—original draft. Yu Wang: Conceptualization, Methodology and Data curation. Henghui Zhang: Methodology, Resources and Data curation. Xinlei Zhao, Hang Su, Dongqing Li and Chen Chen: Investigation, validation, and data curation. Fei Li: Conceptualization, project administration, review & editing.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Competing interests

The authors declare no competing interests.

References

  1. Siegel, R.L., et al., Cancer statistics, 2023. CA Cancer J Clin, 2023. 73(1): p. 17–48.
  2. Sung, H., et al., Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin, 2021. 71(3): p. 209–249.
  3. Dyrskjøt, L., et al., Bladder cancer. Nat Rev Dis Primers, 2023. 9(1): p. 58.
  4. Safiri, S., A.A. Kolahi, and M. Naghavi, Global, regional and national burden of bladder cancer and its attributable risk factors in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease study 2019. BMJ Glob Health, 2021. 6(11).
  5. Catto, J.W.F., et al., Radical Cystectomy Against Intravesical BCG for High-Risk High-Grade Nonmuscle Invasive Bladder Cancer: Results From the Randomized Controlled BRAVO-Feasibility Study. J Clin Oncol, 2021. 39(3): p. 202–214.
  6. Li, Y., et al., Deciphering riddles in molecular subtyping of bladder cancer. Asian J Urol, 2025. 12(2): p. 217–231.
  7. Muñoz-Espín, D. and M. Serrano, Cellular senescence: from physiology to pathology. Nat Rev Mol Cell Biol, 2014. 15(7): p. 482–96.
  8. Schmitt, C.A., B. Wang, and M. Demaria, Senescence and cancer – role and therapeutic opportunities. Nat Rev Clin Oncol, 2022. 19(10): p. 619–636.
  9. Paez-Ribes, M., et al., Targeting senescent cells in translational medicine. EMBO Mol Med, 2019. 11(12): p. e10234.
  10. Gorgoulis, V., et al., Cellular Senescence: Defining a Path Forward. Cell, 2019. 179(4): p. 813–827.
  11. Campisi, J., Senescent cells, tumor suppression, and organismal aging: good citizens, bad neighbors. Cell, 2005. 120(4): p. 513–22.
  12. Wang, L., et al., An Acquired Vulnerability of Drug-Resistant Melanoma with Therapeutic Potential. Cell, 2018. 173(6): p. 1413–1425.e14.
  13. Meyer, K.D., et al., Comprehensive analysis of mRNA methylation reveals enrichment in 3′ UTRs and near stop codons. Cell, 2012. 149(7): p. 1635–46.
  14. Huang, H., H. Weng, and J. Chen, m(6)A Modification in Coding and Non-coding RNAs: Roles and Therapeutic Implications in Cancer. Cancer Cell, 2020. 37(3): p. 270–288.
  15. Xie, H., et al., METTL3/YTHDF2 m(6) A axis promotes tumorigenesis by degrading SETD7 and KLF4 mRNAs in bladder cancer. J Cell Mol Med, 2020. 24(7): p. 4092–4104.
  16. Han, J., et al., METTL3 promote tumor proliferation of bladder cancer by accelerating pri-miR221/222 maturation in m6A-dependent manner. Mol Cancer, 2019. 18(1): p. 110.
  17. Han, S.P., Y.H. Tang, and R. Smith, Functional diversity of the hnRNPs: past, present and perspectives. Biochem J, 2010. 430(3): p. 379–92.
  18. Liu, H., et al., Interaction of lncRNA MIR100HG with hnRNPA2B1 facilitates m(6)A-dependent stabilization of TCF7L2 mRNA and colorectal cancer progression. Mol Cancer, 2022. 21(1): p. 74.
  19. Jiang, F., et al., HNRNPA2B1 promotes multiple myeloma progression by increasing AKT3 expression via m6A-dependent stabilization of ILF3 mRNA. J Hematol Oncol, 2021. 14(1): p. 54.
  20. Li, F., et al., HnRNP-F promotes the proliferation of bladder cancer cells mediated by PI3K/AKT/FOXO1. J Cancer, 2021. 12(1): p. 281–291.
  21. Li, F., et al., HnRNP-F regulates EMT in bladder cancer by mediating the stabilization of Snail1 mRNA by binding to its 3′ UTR. EBioMedicine, 2019. 45: p. 208–219.
  22. Huang, H., et al., Recognition of RNA N(6)-methyladenosine by IGF2BP proteins enhances mRNA stability and translation. Nat Cell Biol, 2018. 20(3): p. 285–295.
  23. Richters, A., K.K.H. Aben, and L. Kiemeney, The global burden of urinary bladder cancer: an update. World J Urol, 2020. 38(8): p. 1895–1904.
  24. Pilié, P.G., et al., State-of-the-art strategies for targeting the DNA damage response in cancer. Nat Rev Clin Oncol, 2019. 16(2): p. 81–104.
  25. Liu, X., et al., DUSP5 promotes osteogenic differentiation through SCP1/2-dependent phosphorylation of SMAD1. Stem Cells, 2021. 39(10): p. 1395–1409.
  26. Chen, H.F., H.C. Chuang, and T.H. Tan, Regulation of Dual-Specificity Phosphatase (DUSP) Ubiquitination and Protein Stability. Int J Mol Sci, 2019. 20(11).
  27. Kutty, R.G., et al., Dual Specificity Phosphatase 5-Substrate Interaction: A Mechanistic Perspective. Compr Physiol, 2017. 7(4): p. 1449–1461.
  28. Han, S.H., et al., Genetic variants reshape the m(6)A epitranscriptome and drive transcriptomic reprogramming in colorectal cancer. Sci Rep, 2025. 15(1): p. 39384.
  29. Liu, K., et al., When therapy-induced senescence meets tumors: A double-edged sword: A review. Medicine (Baltimore), 2025. 104(24): p. e42886.
  30. Kumari, R. and P. Jat, Mechanisms of Cellular Senescence: Cell Cycle Arrest and Senescence Associated Secretory Phenotype. Front Cell Dev Biol, 2021. 9: p. 645593.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top