DAZAP1 promotes cancer progression and chemotherapy resistance by stabilizing PIN1 protein in gastric cancer

Abstract

Gastric cancer (GC) continues to be a fatal disease globally, largely due to the lack of dependable molecular indicators enabling early diagnosis and therapeutic intervention. Single-cell transcriptomic analysis revealed significant enrichment of DAZAP1 in proliferating and malignant gastric epithelial cells. Using a combined analysis of single-cell and bulk RNA-seq datasets, we further recognized DAZAP1 as a putative oncogene correlated with poor clinical outcomes in GC. Functional experiments demonstrated that DAZAP1 promotes tumor proliferation, cell cycle progression, and chemotherapy resistance in vitro and in vivo. Mechanistically, DAZAP1 bound and stabilized USP34 mRNA, leading to increased USP34 protein expression, which in turn mediated the deubiquitination and stabilization of the oncoprotein PIN1. This subsequently resulted in activation of the MAPK signaling pathway, driving GC progression and chemoresistance. Furthermore, we revealed that DAZAP1 expression is post-transcriptionally regulated by m6A modification through the demethylase ALKBH5, which protects DAZAP1 mRNA from YTHDF2-mediated degradation. Collectively, our findings establish the ALKBH5/DAZAP1/USP34/PIN1/MAPK axis as a key regulatory mechanism in gastric tumorigenesis and chemoresistance, underscoring DAZAP1 as a promising candidate for therapeutic and diagnostic applications in GC.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s10565-025-10118-3.

Introduction

Gastric cancer (GC) is the fifth most commonly diagnosed malignancy worldwide and the fourth major contributor to cancer mortality,, ranks third in incidence among malignant tumors in China (Sung et al. 2021; Yang et al. 2023).. Moreover, there is still lacking effective targeted strategies for patients at advanced stages (Xia et al. 2022). Accordingly, it is essential to uncover novel biomarkers that may aid in the diagnosis and management of GC.

Deleted in azoospermia-associated protein 1 (DAZAP1), belonging to the heterogeneous RNA-binding proteins, involves in multiple cellular processes, including alternative splicing, RNA translation, RNA transcription, and RNA degradation (Gerstberger et al. 2014). DAZAP1 is essential in promoting the growth of normal tissues and spermatogenesis (Smith et al. 2011). However, the biological functions of DAZAP1 in malignancies remain poorly understood. Currently, the effects of DAZAP1 on malignancies remain a matter of debate. DAZAP1 has been identified as an oncogene in certain cancers (Wang et al. 2021; Zhou et al. 2022). Yu et al. found a correlation between alternative splicing of DAZAP1 and poor overall survival in pancreatic cancer patients (Yu et al. 2019). In contrast, Chen et al. discovered that DAZAP1 exerted tumor-suppressive effects in esophageal squamous cell carcinoma (ESCC) (Chen et al. 2020). This study represents the first investigation of DAZAP1 expression and its biological functions in GC.

This study is the first to show that DAZAP1 levels are significantly elevated in GC and linked to unfavorable patient outcomes, highlighting its promise as a new prognostic biomarker. Mechanistically, we demonstrate that DAZAP1 binds to and stabilizes USP34 mRNA, promoting USP34 protein expression, which in turn enhances PIN1 protein stability and activates the MAPK signaling pathway, ultimately driving GC progression and chemoresistance. Furthermore, we confirm that DAZAP1 expression is regulated by m6A modification, specifically through the demethylase ALKBH5, which removes m6A marks from DAZAP1 mRNA and counteracts YTHDF2-mediated mRNA degradation, thereby upregulating DAZAP1 expression. These findings not only elucidate the critical role of the ALKBH5/DAZAP1/USP34/PIN1/MAPK signaling axis in gastric tumorigenesis but also underscore its promise as a molecular target for cancer therapy of GC.

Material and methods

Patients and ethical statement

Our study comprised 45 GC patients with no history of anticancer therapy before sample collection.The enrollment of patients was carried out with their informed consent. The Research Ethics Committee of Guangzhou First People’s Hospital approved this study.

Data collection

RNA expression and clinical data were obtained from GEO (GSE26942, GSE14210) and TCGA, with TCGA data converted to TPM. GEO microarray data were quantile-normalized using preprocessCore in R, and mean values were used for genes with multiple probes.

Single-cell analysis

Public scRNA-seq datasets from GEO (GSE134520, GSE163558, GSE167297) (https://www.ncbi.nlm.nih.gov/geo/), and ENA (PRJEB25780, https://www.ebi.ac.uk/ena/browser/home) were analyzed, covering GC and adjacent normal tissues. Raw data were processed into Seurat (v5.3.0) objects in R, with low-quality cells removed. Low-quality cells were removed through a quality control process using the following criteria: cells expressing under 200 or over 6,000 genes, or exhibiting mitochondrial gene proportions greater than 10%, yielding 105,019 high-quality cells. SCTransform was used for normalization and variance stabilization. Principal component analysis based on the 2,000 most variable genes yielded 20 principal components, and cells were clustered into 25 populations by Uniform Manifold Approximation and Projection (UMAP) for Dimension Reduction (Satija et al. 2015), with batch effects corrected via Harmony. Cell identities were determined using well-established markers curated from the CellMarker database and previous publications (Hu et al. 2023). CellCycleScoring determined cell-cycle phases, and CopyKAT inferred large-scale copy number variations (CNVs) to distinguish malignant from non-malignant cells.

Flow cytometry

For cell cycle analysis, 1 × 10⁶ cells were washed with ice-cold PBS, fixed in 75% ethanol at –20 °C overnight, and stained with PI/RNase Staining Buffer (BD, NJ, USA) for 15 min at room temperature before flow cytometry. For apoptosis assays, cells were resuspended in 1X binding buffer, stained with PE-Annexin V and 7-AAD (BD, NJ, USA) for 15 min in the dark, and analyzed by flow cytometry. Data were processed using FlowJo software.

Animal experiment

Male BALB/c nude mice aged 5–6 weeks (Jiangsu Yao Kang Biotech) were obtained for experiments. For xenograft formation, 2 × 10⁶ cells suspended in 100 μL PBS were injected under the skin of the right thigh. Tumor dimensions were recorded every three days, and volumes calculated using the formula: volume (mm³) = length × width² × 0.5. After four weeks, mice were euthanized, and tumors were removed, imaged, and weighed. For chemotherapy experiments, tumor-bearing mice were randomly assigned to four groups (n = 5 per group) one week after tumor implantation: NC treated with PBS, NC treated with oxaliplatin, shDAZAP1 treated with PBS, and shDAZAP1 treated with oxaliplatin. Oxaliplatin was delivered intraperitoneally at 5 mg/kg, three times weekly, according to previously reported methods (Wu et al. 2025). Tumor growth and body weight were monitored every 3 days. At study end, tumors were collected for further analysis.

Drug sensitivity analysis

Half-maximal inhibitory concentration (IC50) values for common chemotherapeutic drugs were calculated using the "pRRophetic" package. Three commonly used chemotherapeutic drugs including 5-Fluorouracil (5-FU), oxaliplatin, and cisplatin (DDP) for gastric cancer were applied to detect the role of DAZAP1 in chemosensitivity by using the MTT assay.

Mass spectrometry

Tumor cells were harvested, broken down, and purified prior to mass spectrometry analysis. Differentially expressed proteins were identified by assessing changes in peptide-to-protein ratios. Tumor cells were harvested, broken down, and purified prior to mass spectrometry analysis. The data were analyzed using the Discovery program to obtain information on master proteins and genes, relative abundance, GO and KEGG pathways, PSM, and other related parameters.

m6A dot blot assay

Total RNA was extracted following the previously described procedure and denatured by heating at 65 °C for 5 min. A mixture of RNA (200 ng and 100 ng) was dissolved in RNase-free water and applied onto a nitrocellulose membrane (Univ, Shanghai, China) using a Bio-Dot system (Bio-Rad, USA). The membrane was then subjected to UV crosslinking, blocked in 5% skim milk, and incubated with m6A primary antibody (1:1000 dilution, Huabio, Hangzhou, China) overnight at 4 °C. Subsequently, Goat anti-Rabbit IgG conjugated to HRP (1:300,000, Huabio, Hangzhou, China) was applied to the membrane. Signal detection was performed using chemiluminescence (Bio-Rad, USA).

m6A RNA immunoprecipitation assay (MeRIP)

The m6A modification of specific transcripts was analyzed using the Methylated RNA Immunoprecipitation (MeRIP) technique, performed with the BersinBio™ MeRIP Kit (BersinBio, Guangzhou, China). RNA was isolated using TRIzol reagent and sheared to approximately 300 bp in length with an ultrasonic disruptor. The resulting RNA fragments were incubated with an anti-N6-methyladenosine antibody at 4 °C for 4 h using a vertical rotator. This was followed by binding with protein A/G beads for an additional 1 h. Afterward, RNA was recovered and cleaned using a phenol/chloroform/isoamyl alcohol mixture (25:24:1). The purified RNA was then reverse-transcribed into cDNA and subjected to qPCR. Details of the primers used are provided in Supplementary Table S1.

RNA immunoprecipitation (RIP) assay

RNA-binding protein enrichment was conducted using the Magna RIP™ Kit (Beyotime, China, #P1801S) following the manufacturer’s instructions. In short, gastric cancer (GC) cells were disrupted in RIP lysis buffer supplemented with protease and RNase inhibitors. The lysates were incubated with magnetic beads pre-coated with anti-DAZAP1 (Proteintech, #11120–1-AP), anti-YTHDF2 (Proteintech, #24744–1-AP), or normal rabbit IgG as a negative control at 4 °C overnight. Following stringent washes, RNA bound to the immunoprecipitates was isolated and purified. The target RNA enrichment (USP34 for DAZAP1, and DAZAP1 for YTHDF2) was assessed via RT-qPCR and normalized against input levels.

Actinomycin D chase assay

To assess mRNA stability, GC cells were treated with 5 μg/mL actinomycin D (MedChemExpress, USA) to suppress new RNA synthesis. In DAZAP1-silenced cells, RNA was collected at 0, 2, 4, and 6 h post-treatment to evaluate the stability of USP34 transcripts. Likewise, in cells with ALKBH5 knockdown, samples were harvested at identical time points to examine DAZAP1 transcript degradation. Total RNA was extracted, and target transcript abundance was quantified by RT-qPCR. mRNA half-lives were estimated by comparing the remaining RNA at each time point relative to the 0-h baseline.

Statistical analysis

All statistical analyses were carried out using R software (v4.1.3) and GraphPad Prism (v8.3). Data are shown as mean ± standard deviation from at least three independent experiments. Group comparisons were conducted via Student’s t-test or ANOVA, depending on the context. Correlation analysis was performed using Spearman’s rank coefficient. Kaplan–Meier survival curves were evaluated with the log-rank test. A P-value below 0.05 was deemed statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Results

Single-cell transcriptomic landscape reveals DAZAP1 enrichment in proliferating and malignant GC cells

To comprehensively determine the expression profile and putative function of DAZAP1 in GC at the single-cell level, we integrated four publicly available single-cell RNA sequencing datasets of GC (Fig. 1A). Upon completion of stringent quality assessment and normalization, all cells were projected into a two-dimensional UMAP space and grouped into 25 distinct clusters based on their transcriptional profiles (Fig. 1B). A dot plot displaying representative marker genes revealed distinct transcriptional identities for each cluster (Fig. 1C). These clusters were subsequently annotated into major cell lineages, including Gastric_mucus-secreting_Cells, Chief_Cells, Enterochromaffin_like_Cells, as well as stromal and immune subpopulations (Fig. 1D–E).

Fig. 1.

Single-cell transcriptomic landscape reveals DAZAP1 enrichment in proliferating and malignant gastric cancer cells. A UMAP visualization of four integrated single-cell gastric cancer datasets. B Unsupervised clustering identifies 25 distinct cellular clusters. C Dot plot displaying expression of representative marker genes across clusters. D Dot plot showing the annotated cell-type identities based on canonical markers. E UMAP plot showing distribution of annotated cell subpopulations. F Vlnplot visualization of DAZAP1 expression, showing strong enrichment in proliferating cancer cells. G Cell-cycle phase analysis revealing that DAZAP1 expression peaks in S and G2/M phases. H CopyKAT analysis showing inferred CNV profiles in gastric cancer cells. I Dotplot demonstrating preferential enrichment of DAZAP1 in cells with high CNV (malignant) signatures

Notably, DAZAP1 expression was markedly enriched in the Proliferating_Cancer_Cell population (Fig. 1F), suggesting a potential role in tumor cell proliferation. Cell-cycle analysis further showed that DAZAP1 expression peaked during the S and G2/M phases (Fig. 1G), indicating a strong association with active DNA replication and mitotic activity. To explore whether DAZAP1 expression correlates with malignant transformation, we performed CopyKAT analysis to infer copy number variations (CNVs) from single-cell transcriptomes (Fig. 1H). The results revealed that DAZAP1 was preferentially expressed in cells with high CNV burden, representing genomically unstable and malignant populations (Fig. 1I).

Consistent with our previous study, bulk RNA-seq analysis of multiple independent cohorts confirmed that DAZAP1 shows pronounced upregulation in GC tissues relative to normal mucosal counterparts. (Fig. S1A–C). Quantitative RT-PCR and WB analyses further validated that DAZAP1 expression is elevated at both the mRNA and protein levels in GC cell lines (Fig. S1D–F). DAZAP1 expression in tumor tissues was prominently elevated as shown by immunohistochemical analysis. (Fig. S1G–H), Survival analysis revealed that patients with higher DAZAP1 expression experienced shorter overall survival. (Fig. S1I). Collectively, these results identify DAZAP1 as a consistently upregulated gene in GC that may promote tumor proliferation and malignant progression. Its strong association with poor clinical outcomes indicates that DAZAP1 could act as both a potential driver and biomarker of GC malignancy.

Inhibition of DAZAP1 suppresses GC proliferation and induces apoptosis

To understand how DAZAP1 influences GC tumorigenesis and development on a cellular level, we generated AGS-sh-DAZAP1 and BGC823-sh-DAZAP1 cell lines by applying DAZAP1-shRNA to suppress DAZAP1 expression. Western blotting analysis verified that DAZAP1 expression levels were stably repressed in AGS-sh-DAZAP1 and BGC823-sh-DAZAP1 cell lines (Fig. 2A). We discovered that DAZAP1 knockdown decreased the proliferation capacity of AGS and BGC823 cells (Fig. 2B-C). Consistent with this, colony formation assay (Fig. 2D) and EDU staining assay (Fig. 2E) demonstrated that DAZAP1 knockdown significantly inhibited GC cell proliferation, suggesting that suppressing DAZAP1 expression restrains gastric cancer cell growth. To elucidate the oncogenic mechanism of DAZAP1, we utilized cell cycle and apoptosis analyses. Flow cytometry analysis showed that DAZAP1 knockdown result in a higher proportion of cells in the G1 phase and fewer cells in the S and G2 phases., resulting in cell cycle arrest and proliferation inhibition (Fig. 2F). Moreover, DAZAP1 knockdown elevated the apoptosis rate (Fig. 2G). Furthermore, the impact of DAZAP1 on GC growth was investigated through in vivo experiments. Mice injected with BGC823-sh-DAZAP1 cells exhibited significantly smaller tumor sizes and slower growth rates (Fig. 2H-I). Following animal sacrifice, tumor tissues were harvested and weighed, and DAZAP1 knockdown resulted in a marked decrease in tumor weight. (Fig. 2J). These results suggest that DAZAP1 has a direct influence on GC tumor growth, strengthening its candidacy as a therapeutic target in GC treatment.

Fig. 2.

DAZAP1 knockdown suppresses proliferation and induces apoptosis in gastric cancer. A DAZAP1 knockdown efficiency was validated by Western blot analysis. B-C Cell proliferation was assessed by MTT assay in AGS and BGC823 cells following DAZAP1 knockdown. D Colony formation assay revealed the inhibitory effect of DAZAP1 knockdown on GC cell proliferative capacity; accompanying histograms quantify the number of colonies. E Analysis of cell proliferation by EdU assay. GC cells were stained with EdU (proliferating cells, cyan) and Hoechst (nuclei, blue), and imaged by fluorescence microscopy (Olympus). The bar graph quantifies the proliferation rates across different GC cell lines. Data are presented as mean ± SD from at least three independent experiments. F Cell cycle distribution in DAZAP1-knockdown GC cells was analyzed by flow cytometry; histograms show the proportion of cells in G1, S, and G2/M phases. G Apoptosis assay indicating that DAZAP1 knockdown promotes GC cell apoptosis; the histogram quantifies the percentage of apoptotic cells. H Representative photographs of subcutaneous tumor xenografts from nude mice injected with control or DAZAP1-knockdown GC cells. I Tumor growth curves were generated based on volume measurements taken every three days. Data are shown as mean ± SD (n = 5 mice per group). J Terminal tumor weights were compared across the experimental groups. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Elevated DAZAP1 enhances GC proliferation and cell cycle progression

To further investigate the impact of DAZAP1 on the progression of GC, we also developed DAZAP1 overexpression cells. WB revealed that DAZAP1 was overexpressed in BGC823-oe-DAZAP1 and AGS-oe-DAZAP1 cell lines (Fig. 3A). MTT experiment (Fig. 3B-C), colony formation assay (Fig. 3D), and the EdU assay (Fig. 3E) revealed that the overexpression of DAZAP1 significantly augmented the proliferation capacity of GC cells. Flow cytometry analysis showed that DAZAP1 overexpression significantly increased the proportion of cells in the G2 phase (Fig. 3F). Furthermore, in vivo experiments unveiled that mice injected with BGC823-oe-DAZAP1 cells exhibited a substantial augmentation in both tumor size and growth rate, in comparison to animals injected with control cells (Fig. 3G-H). The DAZAP1 overexpression group displayed a statistically significant elevation in tumor weight (Fig. 3I). Collectively, the results suggest that DAZAP1 is a pivotal mediator of GC progression.

Fig. 3.

DAZAP1 promotes the proliferation of gastric cancer cells. A Overexpression of DAZAP1 was confirmed by Western blot. B-C Proliferation of gastric cancer cells following DAZAP1 overexpression was assessed by MTT assay. D Colony formation assay evaluating the effect of DAZAP1 overexpression on GC cell proliferation. Histograms quantify the number of colonies after crystal violet staining. E EdU staining of proliferating cells in control and DAZAP1-overexpressing groups. Nuclei were stained with Hoechst (blue), and EdU/Hoechst-positive proliferating cells are shown in cyan. Histograms quantify proliferation rates across different GC cell lines. F Cell cycle distribution in GC cells analyzed by flow cytometry. Histograms show the proportion of cells in different cell cycle phases. Data represent mean ± SD from three independent experiments. G Representative images of subcutaneous tumor xenografts in nude mice injected with control or DAZAP1-overexpressing cells. H Tumor growth curves measured every 3 days. Data are presented as mean ± SD (n = 5 mice per group). I Comparison of tumor weights among different mouse groups. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

DAZAP1 contributes to multiple chemotherapy resistance in GC

Extensive research has shown that drug resistance is the biggest barrier to chemotherapy effectiveness in cancer patients and often leads to poor prognosis in patients (Vasan et al. 2019). Analysis of the GSE14210 dataset showed that DAZAP1 expression was significantly upregulated in the chemotherapy-resistant group, suggesting an association between DAZAP1 expression and chemoresistance (Fig. 4A). To further validate this, we employed the calcPhenotype algorithm to analyze the TCGA_STAD dataset and evaluated the relationship between DAZAP1 expression and sensitivity to commonly used chemotherapeutic agents in gastric cancer, including 5-FU, oxaliplatin, and cisplatin. The results indicated that samples with high DAZAP1 expression exhibited significantly reduced sensitivity to these drugs (Fig. 4B). In cellular experiments, treatment with 5-FU, oxaliplatin, and cisplatin markedly reduced the viability of DAZAP1-knockdown AGS and BGC823 cells (Fig. 4C-D). Flow cytometry analysis further confirmed that DAZAP1 knockdown significantly enhanced 5-FU-induced cell death (Fig. 4E), whereas DAZAP1 overexpression substantially attenuated the pro-apoptotic effect of 5-FU (Fig. 4F). To exclude alternative mechanisms underlying DAZAP1-mediated chemoresistance, we examined the expression of classical drug resistance–related genes, including ABCB1 (P-gp), ABCG2 (BCRP), ABCC1 (MRP1), and DNA repair–associated enzymes ERCC1, BRCA1/2, and MGMT in DAZAP1-knockdown gastric cancer cells. qRT-PCR analyses revealed that DAZAP1 silencing did not significantly alter the mRNA levels of these molecules (Figure S2A), indicating that DAZAP1-mediated chemoresistance is not attributed to canonical efflux transporter or DNA repair pathways. To evaluate the in vivo relevance of these findings, we established a nude mouse xenograft model. Tumor growth was markedly suppressed by the combination of DAZAP1 knockdown and oxaliplatin treatment versus controls (Fig. 4G-H). Consistently, the combination group exhibited significantly lower tumor weight than all other groups (Fig. 4I), confirming that DAZAP1 silencing enhances chemosensitivity in vivo.

Fig. 4.

DAZAP1 promotes multidrug resistance in gastric cancer. A DAZAP1 expression is significantly upregulated in chemoresistant patients compared to chemosensitive patients in the GSE14210 dataset. B High DAZAP1 expression correlates with reduced sensitivity to 5-FU, oxaliplatin, and cisplatin in the TCGA-STAD cohort (calcPhenotype analysis). C-D Viability of DAZAP1-knockdown and control gastric cancer cells following treatment with 5-FU, oxaliplatin, and cisplatin, as measured by MTT assay. E Flow cytometric analysis of apoptosis in DAZAP1-knockdown AGS cells after 5-FU treatment. F Flow cytometric analysis of apoptosis in DAZAP1-overexpressing BGC823 cells after 5-FU treatment. G Representative images of subcutaneous tumor xenografts from nude mice in different treatment groups. H Tumor growth curves showing the inhibitory effect of DAZAP1 knockdown combined with oxaliplatin treatment. I Comparison of final tumor weights among the four experimental groups. Data in C-F, H, I are presented as mean ± SD (n = 3 for in vitro experiments; n = 6 for in vivo experiments). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Enhanced stability of PIN1 protein by DAZAP1 activates the MAPK signaling pathway in GC

To assess the downstream potential targets of DAZAP1 in GC, To determine DEGs in the oe-DAZAP1 and sh-DAZAP1 groups, mass spectrometry analysis was performed. The results showed that 30 genes were significantly downregulated when DAZAP1 was knocked down and upregulated when DAZAP1 was overexpressed (Figs. 5A–B). Given that PIN1 is frequently overexpressed in cancers and contributes to tumorigenesis (Wu et al. 2022), we selected it as a candidate target of DAZAP1. DAZAP1 overexpression increased PIN1 protein levels, whereas DAZAP1 silencing decreased them (Fig. 5C). To investigate how DAZAP1 regulates PIN1, we assessed protein stability using the translation inhibitor cycloheximide. The half-life of PIN1 was shortened in DAZAP1-knockdown cells but prolonged in DAZAP1-overexpressing cells (Fig. 5D). Treatment with the proteasome inhibitor MG132 partially rescued PIN1 degradation induced by DAZAP1 knockdown (Fig. 5E), suggesting that DAZAP1 stabilizes PIN1 by suppressing ubiquitin-mediated proteasomal degradation. Rescue experiments showed that PIN1 siRNA partially reversed the pro-proliferative and chemoresistance phenotypes promoted by DAZAP1 overexpression (Fig. 5F-G). Furthermore, PIN1 inhibition significantly increased apoptosis (Fig. 5H), supporting that DAZAP1 exerts its oncogenic effects at least partly through PIN1 upregulation. KEGG pathway analysis revealed the MAPK signaling pathway as the most significantly altered in DAZAP1-knockdown cells (Fig. 5I). Since PIN1 is known to activate the Raf/MEK/ERK cascade in melanoma (Zhou and Lu 2016), we asked whether DAZAP1-mediated proliferation involves MAPK signaling. DAZAP1 knockdown in BGC823 cells reduced both total and phosphorylated ERK1/2 protein levels, while DAZAP1 overexpression specifically elevated p-ERK1/2 without altering total ERK1/2 (Fig. 5J). Together, these findings demonstrate that DAZAP1 promotes GC progression by enhancing PIN1 protein stability and activating the MAPK signaling pathway.

Fig. 5.

DAZAP1 stabilizes PIN1 protein and activates the MAPK signaling pathway in gastric cancer. A Venn diagram identifying 30 overlapping genes that were upregulated in AGS-oe-DAZAP1 cells and downregulated in both AGS-sh-DAZAP1 lines. B Heatmap displaying expression patterns of the 30 differentially expressed genes. C Western blot analysis of PIN1 protein levels in BGC823 cells with DAZAP1 knockdown or overexpression. Relative expression was quantified and normalized to GAPDH. D PIN1 protein half-life in AGS cells after CHX treatment (100 µmol/L). E Western blot analysis of PIN1 expression in cells pretreated with 10 μM MG132 before CHX addition. DMSO served as the vehicle control. (F-G) MTT assays measuring proliferation F and drug resistance G in BGC823 cells. H Apoptosis rates in PIN1-silenced BGC823-oe-DAZAP1 cells were measured by flow cytometry. I KEGG pathway enrichment analysis in AGS-sh-DAZAP1 versus control cells. J Western blot analysis of ERK1/2 and phosphorylated ERK1/2 (p-ERK1/2) levels in BGC823 cells with modulated DAZAP1 expression. All quantitative data are presented as mean ± SD from three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

DAZAP1 stabilizes PIN1 through an RNA-mediated USP34-dependent mechanism

To elucidate the molecular mechanism by which DAZAP1 stabilizes PIN1, we first examined the ubiquitination level of PIN1 in GC cells. Co‑immunoprecipitation assays following proteasome inhibitor MG132 treatment showed that DAZAP1 knockdown markedly enhanced PIN1 polyubiquitination, indicating that DAZAP1 suppresses ubiquitin‑mediated degradation of PIN1 (Fig. 6A). Our previous study demonstrated that DAZAP1 functions as an RNA‑binding protein (Zhang et al. 2025). Building on this, catRAPID bioinformatic analysis predicted USP34 mRNA as a potential binding target of DAZAP1 (Fig. 6B). Subsequent RNA immunoprecipitation (RIP) assays confirmed the direct interaction between DAZAP1 and USP34 mRNA (Fig. 6C). Actinomycin D chase assays revealed that DAZAP1 knockdown substantially diminished the stability of USP34 mRNA (Fig. 6D). Consistent with this, DAZAP1 knockdown downregulated USP34 expression at both the mRNA and protein levels (Figs. 6E and 6 F). Notably, overexpression of USP34 effectively rescued the decrease in PIN1 protein levels induced by DAZAP1 knockdown (Fig. 6G).

Fig. 6.

DAZAP1 stabilizes PIN1 by enhancing USP34 mRNA stability. A Polyubiquitination of PIN1 analyzed by immunoprecipitation and western blot in MG132-treated GC cells with DAZAP1 knockdown. B Prediction of USP34 mRNA as a DAZAP1-binding target using catRAPID. C RIP-qPCR validation of the direct binding between DAZAP1 and USP34 mRNA. D USP34 mRNA stability measured by actinomycin D chase assay. E, F USP34 expression at the mRNA E and protein F levels upon DAZAP1 knockdown. G PIN1 protein levels rescued by USP34 overexpression in DAZAP1-deficient cells. Data are shown as mean ± SD from ≥ 3 experiments. *P < 0.05, **P < 0.01, ***P < 0.001

In summary, this study demonstrates that DAZAP1, as an RNA‑binding protein, directly binds to and stabilizes USP34 mRNA, thereby maintaining USP34 protein expression and promoting PIN1 deubiquitination and stability. The newly identified DAZAP1–USP34–PIN1 regulatory axis provides novel molecular insights into the post‑transcriptional regulation of protein stability in gastric cancer cells.

ALKBH5 enhances DAZAP1 expression through m6A demethylation

Li et al. certificated that SOX2 exerts oncogenic effects by METTL3-catalyzed methylation in CRC (Li et al. 2019). However, whether and how DAZAP1 expression is regulated by m6A modification remains unclear. To investigate this issue, we treated AGS and BGC823 cells with DAA, a demethylase activator. RT-qPCR analysis revealed that DAA treatment significantly increased DAZAP1 mRNA levels (Fig. 7A), suggesting that DAZAP1 expression may be positively regulated by demethylase activity. Analysis of the TCGA-STAD cohort further showed a strong positive correlation between DAZAP1 and ALKBH5 expression (Fig. 7B). We therefore knocked down ALKBH5 using siRNA in AGS, MKN45, and BGC823 cells (Fig. 7C, Supplementary Fig.  3 A), and observed a corresponding decrease in DAZAP1 protein levels (Fig. 7D, Supplementary Fig. 3B). m6A dot blot assays indicated a marked increase in global m6A levels upon ALKBH5 knockdown (Fig. 7E). Using the SRAMP online tool, we identified three high-confidence m6A modification sites on DAZAP1 mRNA (Fig. 7F) and designed specific primers targeting these regions (Fig. 7G). MeRIP-qPCR confirmed that ALKBH5 knockdown significantly enhanced m6A modification on DAZAP1 mRNA (Fig. 7H). Actinomycin D assays demonstrated that ALKBH5 knockdown accelerated DAZAP1 mRNA decay (Fig. 7I), indicating that m6A modification promotes its mRNA instability. YTHDF2, an m6A reader protein, has been widely reported to recognize m6A-modified transcripts and facilitate their degradation. By integrating POSTAR3 CLIP-seq data and RIP-qPCR results, we confirmed that YTHDF2 directly binds to DAZAP1 mRNA (Figs. 7J, Table S3).

Fig. 7.

ALKBH5 upregulates DAZAP1 expression by erasing m6A modification in gastric cancer. A DAZAP1 mRNA levels in AGS and BGC823 cells treated with the demethylase activator DAA, measured by RT‑qPCR. B Correlation analysis between DAZAP1 and ALKBH5 expression in the TCGA‑STAD cohort. C ALKBH5 knockdown efficiency in AGS and MKN45 cells transfected with siRNA, assessed by RT‑qPCR. D DAZAP1 protein expression in ALKBH5‑knockdown GC cells, analyzed by Western blot. E Dot blot analysis of global m6A modification levels in control and ALKBH5-knockdown BGC823 cells. F Schematic of predicted m6A modification sites on DAZAP1 mRNA using the SRAMP algorithm. G Design of specific primers targeting the m6A-modified regions of DAZAP1 mRNA. H MeRIP‑qPCR analysis of m6A enrichment on DAZAP1 mRNA in control and ALKBH5‑knockdown cells. I DAZAP1 mRNA stability assessed by actinomycin D chase assay in ALKBH5‑deficient cells. J RIP‑qPCR validation of YTHDF2 binding to DAZAP1 mRNA in GC cells. Data are presented as mean ± SD from at least three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001

In summary, this study systematically reveals that ALKBH5 removes m6A marks from DAZAP1 mRNA, thereby counteracting YTHDF2-mediated mRNA degradation and ultimately upregulating DAZAP1 expression. The identification of the ALKBH5–m6A–YTHDF2–DAZAP1 regulatory axis sheds light on the mechanism of the post-transcriptional regulation of DAZAP1 in GC.

Discussion

GC continues to pose a major global health burden and is predictive of poor clinical outcome. and difficulties in effective clinical intervention (Smyth et al. 2020). The presence of intratumoral heterogeneity in advanced GC further complicates the implementation of single chemotherapy regimens or targeted therapies, limiting their efficacies (Song et al. 2017). Although previous studies have implicated DAZAP1 in tumorigenesis and disease progression (Wang et al. 2021), full elucidation of its function remains largely unavailable. Hence, we aimed to address this critical knowledge gap using an in-depth examination of the novel gene DAZAP1 in the context of GC.

DAZAP1 is recognized as a multifunctional RBP predominantly localized within the nucleus, playing crucial regulatory roles in diverse tumor types. Choudhury et al. showed the integration of splicing control by DAZAP1 into MEK/ERK-regulated processes related to proliferation and migration in non-small cell lung carcinoma (Choudhury et al. 2014). Deng et al. also found that DAZAP1 is highly expressed in hepatic carcinoma and promoted liver carcinoma cell proliferation, migration, and invasion (Deng et al. 2022). Indeed, the existing literatures have limited researches on the role of DAZAP1 in the pathogenesis of GC. Our study indicates that DAZAP1 is significantly upregulated in GC and is critically involved in driving tumor progression. Specifically, DAZAP1 appears to inhibit cell apoptosis and accelerate the cell cycle, contributing to increased tumor growth and aggressiveness. Furthermore, the study suggests that high expression of DAZAP1 is associated with a poorer prognosis in GC patients. These findings shed light on the potential importance of DAZAP1 as a therapeutic target and prognostic marker in GC. An important observation is that the silencing of DAZAP1 demonstrated a sensitizing effect on various chemotherapy drugs. Collectively, these data support the notion that targeting DAZAP1 could be a promising therapeutic approach for patients facing chemotherapy resistance in GC.

We subjected to further analysis the molecular basis of DAZAP1 in enhancing GC progression. Based on previous reports demonstrating that USP34 stabilizes PIN1 in glioma stem cells (Zhu et al. 2024), we investigated whether DAZAP1 regulates PIN1 through USP34. Mechanistically, we discovered that DAZAP1 directly binds to and stabilizes USP34 mRNA, promoting USP34 protein expression. As a deubiquitinating enzyme, USP34 effectively reduces PIN1 ubiquitination and subsequent degradation, thereby maintaining PIN1 protein stability. These findings reveal a novel post-translational regulatory pathway through which DAZAP1 enhances PIN1 stability via USP34 mediation. Extensive researches have substantiated the pivotal roles of PIN1 in governing diverse cellular processes such as cell cycle progression, cellular motility, and apoptosis (Pan et al. 2010). Furthermore, augmented expression of PIN1 had been consistently observed in malignant cells, highlighting its correlation with aggressive phenotypes (Zheng et al. 2009; Liang et al. 2019). Importantly, PIN1 had been implicated in augmenting oncogenic functions while suppressing tumor suppressors, further driving multiple signaling pathways associated with cancer biology (Wu et al. 2022). Hence, PIN1 is viewed as a significant regulator of malignant processes. Notably, the upregulation of PIN1 expression initiated the phosphorylation of ERK, activated cancer-promoting signaling pathways, and ultimately promoted the proliferation of GC cells. The rescue experiment showed that knocking down PIN1 reverse tumor progression and chemotherapy resistance induced by DAZAP1. This finding suggests that PIN1 acts as a critical downstream effector of DAZAP1 in promoting GC progression and conferring resistance to chemotherapy.

The underlying causes for the upregulated expression of DAZAP1 in GC are yet to be fully clarified. Over the past few years, RNA m6A modification has been recognized as an important regulator in epigenetics, influencing numerous cellular activities, including mRNA maturation and degradation (Zhang et al. 2020a). Extensive investigations have explored the functional implications of m6A modification in various cancer types, such as leukemia, liver cancer, colon cancer, and breast cancer (Zhang et al. 2022, 2020b; Yang et al. 2020; Wan et al. 2022). In light of this, our study aims to delve into the intricate relationship between DAZAP1 expression in cancer and m6A modification. Notably, we made an intriguing discovery that inhibiting the overall level of m6A methylation resulted in an upregulation of DAZAP1 expression. Remarkably, we found that ALKBH5 exerted a positive regulatory effect on DAZAP1 expression, influencing the m6A modification of DAZAP1 mRNA. These compelling findings strongly suggest that ALKBH5-mediated m6A methylation may be fundamental to maintaining the expression of DAZAP1.

It is worth noting that while our study has systematically elucidated the ALKBH5/DAZAP1/USP34/PIN1/MAPK regulatory axis, several aspects warrant further investigation. The precise structural basis of DAZAP1-USP34 mRNA interaction remains to be fully characterized, which would provide deeper insights into the RNA recognition mechanism. Additionally, the potential involvement of other m6A readers in fine-tuning DAZAP1 expression represents an interesting direction for future research.

Our findings reveal a novel epitranscriptional regulatory mechanism in gastric cancer progression, where ALKBH5, a key m6A demethylase, removes m6A methylation marks from DAZAP1 transcripts, thereby impairing recognition and binding by the m6A reader protein YTHDF2. Since YTHDF2 is known to promote the degradation of target mRNAs (Du et al. 2016; Lee et al. 2020), ALKBH5-mediated demethylation consequently stabilizes DAZAP1 mRNA and enhances its protein expression, ultimately driving gastric cancer progression. The accumulated DAZAP1 protein further activates the downstream DAZAP1/USP34/PIN1 signaling axis, which not only promotes MAPK pathway activation but also confers resistance to conventional chemotherapeutic agents including 5-fluorouracil, oxaliplatin, and cisplatin. These findings establish DAZAP1 as a critical molecular node connecting m6A epitranscriptional regulation to malignant progression in GC, highlighting its dual utility serving as a prognostic marker and a candidate for therapeutic intervention. Further investigation into this comprehensive regulatory network may provide innovative combination strategies to overcome chemotherapy resistance in advanced GC, particularly through co-targeting the ALKBH5/DAZAP1 axis alongside conventional chemotherapeutic regimens.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Lin Lu conceived and supervised the study. Yong Wu contributed to experimental design and critically revised the manuscript. Xisheng Fang and Chengyin Weng participated in research conceptualization. Peiling Zhang, Lujuan Ma, and Yitian Wei performed experiments and analyzed data; Peiling Zhang also prepared the figures. Qian Peng assisted in experimental work and manuscript drafting. Hong Xiang provided methodological support. All authors reviewed and approved the final manuscript.

Funding

This research was supported by the Natural Science Foundation of Guangdong Province, China (No. 2021A1515011113), the Guangzhou Planned Project of Science and Technology (2023A03J0966), and Special Funds for the Cultivation of Guangdong College Students' Scientific and Technological Innovation (pdjh2025bk023).

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Publicly available datasets used in this study can be accessed through their respective repositories as referenced in the manuscript.

Declarations

Ethical approval

The GC patients in our study were diagnosed with pathologically confirmed gastric adenocarcinoma through surgical resection, and complete follow-up information was available for all cases. The acquisition of human tissue specimens for this study approved by the Ethics Committee of Guangzhou First People's Hospital (Approval No: K-2021–140), with informed consent obtained from all donors. All animal experiments were conducted in accordance with the approval from the Laboratory Animal Ethics Committee, School of Medicine, South China University of Technology. The approval date for this research protocol was August 24, 2020.

Conflicts of interest

The authors declare no competing interests.