DAZAP1 maintains gastric cancer stemness by inducing mitophagy

Abstract

Stem cells play a pivotal role in the malignant behavior of gastric cancer (GC), complicating its treatment and prognosis. However, the regulatory mechanisms of GC stem cells (GCSCs) remain poorly understood. DAZ-associated protein 1 (DAZAP1), a splicing regulator linked to various malignancies, has an unclear role in GC. This study investigated DAZAP1's impact on GC stemness and its mechanisms. DAZAP1 promoted tumor progression in GCSCs, as shown by sphere formation assays and stemness marker analysis. Functional enrichment analysis suggested that DAZAP1 enhanced tumor stemness by promoting oxidative phosphorylation (OXPHOS), which was validated through Seahorse assays and measurements of mitochondrial potential. Transmission electron microscopy and immunofluorescence analyses demonstrated that DAZAP1 promoted mitophagy. RNA immunoprecipitation and PCR analysis revealed that DAZAP1 regulated the splicing and expression of the mitophagy-related gene ULK1 through nonsense-mediated mRNA decay. Rescue experiments showed that overexpression of ULK1 reversed the suppression of GC stemness and OXPHOS levels induced by DAZAP1 silencing. Our findings indicate that DAZAP1 reduces ULK1 decay, thereby activating mitophagy and enhancing OXPHOS to fulfill the metabolic demands of cancer stem cells. These findings highlight the therapeutic potential of DAZAP1 as a target for treating GC.

Keywords: Autophagy; Cell biology; Gastric cancer; Mitochondria; Oncology; Stem cells.

Figure 1. DAZAP1 expression and its association with prognosis in GC.

(A) Integration of 3 scRNA-seq datasets (GSE167297, GSE183904, and GSE206785) yielded gene expression profiles from 280,130 cells across 58 primary GC samples and 39 normal tissues. (B and C) Cells were grouped into 37 clusters and further annotated into 11 cell types using the KNN algorithm. (D and E) DAZAP1 was predominantly expressed in epithelial cells, showing higher levels in malignant cells compared with normal tissues (P < 2 × 10^–16). (F and G) Elevated DAZAP1 expression was observed in 6 GC cell lines relative to normal gastric mucosal epithelial cells. (H–J) IHC confirmed elevated DAZAP1 expression in GC (n = 52) compared with normal tissue (n = 20) (H and I), which correlated with poorer survival outcomes (n = 52) (J). Scale bars: 300 μm (top) and 200 μm (bottom). Data are presented as mean ± SD. Statistical analysis was by Wilcoxon’s rank-sum test (E), 1-way ANOVA followed by Tukey’s HSD post hoc test for multiple comparisons (F), log-rank (Mantel-Cox) test (I), or by unpaired Student’s t test (J). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2. Elevated DAZAP1 expression in GCSCs.

(A) Stemness indices of TCGA-STAD samples, calculated using the OCLR algorithm, were higher in the DAZAP1 high-expression group.(B) CytoTRACE analysis revealed higher DAZAP1 expression in high-stemness tumor cells (P < 2 × 10^–16). (C) DAZAP1 expression was positively correlated with CytoTRACE scores in tumor epithelial cells (r = 0.38, P < 2 × 10^–16). (D) Pseudotime analysis indicated that DAZAP1 expression increased progressively from undifferentiated to differentiated states. (E) DAZAP1 expression was positively correlated with the stem cell markers EPCAM and CD44. (F and G) Spheroid cultures enriched for GCSCs exhibited elevated levels of stemness genes and DAZAP1 mRNA compared with adherent cells. (H) Western blot analysis confirmed increased DAZAP1 protein levels in spheroid cells. Data are presented as mean ± SD. Statistical analysis was by unpaired Student’s t test (A, F, and G), Wilcoxon’s rank-sum test (B), or Spearman’s correlation analysis (C and E). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 3. The impact of DAZAP1 on malignant phenotypes in GC.

(A) DAZAP1 knockdown in AGS and NCI-N87 cells reduced its expression, while overexpression in HGC27 cells elevated DAZAP1 levels. EV, empty vector. (B) CCK8 assays demonstrated decreased cell viability with DAZAP1 knockdown and increased cell proliferation with its overexpression. (C) EdU assays confirmed that DAZAP1 enhances cell proliferation. (D) Wound healing assays demonstrated that DAZAP1 knockdown reduces cell motility, while DAZAP1 overexpression increases it. (E) Transwell assays showed that DAZAP1 knockdown significantly inhibited, while DAZAP1 overexpression enhanced, the migration and invasiveness of GC cells. (F–H) In vivo tumor formation assays with NCI-N87 cells revealed that DAZAP1 knockdown inhibited tumor growth rate and tumor weight (n = 5/group). Data are presented as mean ± SD. Statistical analysis was by 2-way ANOVA followed by Tukey’s HSD post hoc test for multiple comparisons (B and G), unpaired Student’s t test for comparisons between 2 groups and 1-way ANOVA followed by Tukey’s HSD post hoc test for comparisons among 3 groups (C–E), or 1-way ANOVA followed by Tukey’s HSD post hoc test (H). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 4. DAZAP1 promotes stemness properties in GC.

(A and B) Sphere formation assays indicated that DAZAP1 knockdown reduced the number of spheres formed, while overexpression increased sphere-forming ability. (C) Extreme limiting dilution assay (ELDA) indicated that DAZAP1 knockdown decreases self-renewal capacity, while DAZAP1 overexpression enhances it. (D and E) ALDEFLUOR assays showed a decrease in ALDH-high cells with DAZAP1 knockdown and an increase with DAZAP1 overexpression. (F and G) Quantitative PCR (qPCR) and Western blot analyses revealed that DAZAP1 overexpression upregulated stemness markers SOX2, OCT4, and NANOG, while knockdown downregulated these markers. (H and I) In vivo ELDA using DAZAP1-knockdown NCI-N87 cells demonstrated a substantially reduced tumorigenic capacity, requiring higher cell numbers to form detectable tumors compared with control cells. All quantitative data are presented as the mean ± SD of at least 3 independent experiments. Statistical analysis was by unpaired Student’s t test for comparisons between 2 groups and 1-way ANOVA followed by Tukey’s HSD post hoc test for comparisons among 3 groups (B, E, and F). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 5. DAZAP1 enhances OXPHOS activity and GC cell stemness.

(A and B) KEGG functional enrichment analysis revealed that differentially expressed genes were markedly involved in OXPHOS processes following DAZAP1 knockdown or overexpression. (C) Measurement of ATP content indicated reduced ATP levels with DAZAP1 knockdown and increased levels with overexpression. (D) Seahorse XF Analyzer data confirmed that DAZAP1 knockdown decreased OXPHOS activity, as evidenced by a reduction in OCR. (E) Specific inhibitor and substrate assays revealed suppressed basal OCR, spare respiratory capacity, and ATP production in DAZAP1-knockdown cells, with no change in proton leak. PL, proton leak; MR, maximal respiration; SRC, spare respiratory capacity; NMOC, non-mitochondrial oxygen consumption. (F and G) DAZAP1 overexpression increased basal OCR, proton leak, spare respiratory capacity, and ATP production. (H–J) JC-1 assay demonstrated decreased mitochondrial membrane potential in DAZAP1-knockdown cells, indicated by a lower red/green fluorescence ratio. Scale bars: 50 μm. (K) Gene expression analysis showed a decrease in the expression of OXPHOS complex subunit genes in DAZAP1-knockdown cells and an increase in these genes with overexpression. (L and M) Sphere formation assays demonstrated that the OXPHOS inhibitor Gboxin weakened the sphere-forming capacity enhanced by DAZAP1 overexpression. (N) Western blot analysis revealed that the elevation of stemness markers (OCT4, NANOG, and SOX2) induced by DAZAP1 overexpression was counteracted by the OXPHOS inhibitor Gboxin. Quantitative data are expressed as the mean ± SD from a minimum of 3 independent experiments. Statistical analysis by unpaired Student’s t test (HGC27-EV vs. HGC27-OE) for comparisons between 2 groups and 1-way ANOVA followed by Tukey’s HSD post hoc test (AGS-shNC vs. AGS-sh1 and AGS-sh2) for comparisons among 3 groups (C), 1-way ANOVA followed by Tukey’s HSD post hoc test (E, J, and M), or unpaired Student’s t test (G and K). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 6. DAZAP1 enhances OXPHOS and maintains cell stemness in GC by inducing mitophagy.

(A) TEM images show an increased number of mitophagosomes in DAZAP1-overexpressing (DAZAP1-OE) cells, indicating induced mitophagy. Scale bars: 1 μm (left) and 500 nm (right). (B) Confocal microscopy using MitoTracker Red reveals mitochondrial fragmentation and a reduced mitochondria-to-nucleus ratio following DAZAP1 knockdown (n = 3). (C) Immunofluorescence (IF) analysis shows increased colocalization of mitochondria with LC3B in DAZAP1-OE cells, indicating enhanced mitophagy. (D) DAZAP1 overexpression results in increased colocalization of LC3B with lysosomes, promoting autophagosome-lysosome fusion. (E) Mito-Keima plasmid labeling demonstrates increased mitophagy in DAZAP1-OE cells. Scale bars (C–E): 10 μm. (F) MitoTracker Deep Red (MTDR) dye analysis confirms impaired mitophagy in DAZAP1-knockdown cells and enhanced mitophagy in DAZAP1-OE cells. (G) Oxygen consumption rate (OCR) analysis shows reduced basal OCR, proton leak, and ATP production in GC cells treated with the autophagy inhibitor chloroquine (CQ) and the mitophagy inhibitor Mdivi-1, indicating the role of mitophagy in regulating OXPHOS. (H) Gene expression analysis reveals that Mdivi-1 restores the expression of OXPHOS complex subunit genes (UQCRC1, UQCRC2, SDHA, SDHB, ATP5F1A, ATP5F1B, and NDUFA1) elevated by DAZAP1 overexpression. (I and J) Sphere formation assay shows that Mdivi-1 treatment substantially reduces the sphere-forming ability of DAZAP1-OE cells. (K) Western blot analysis demonstrates that Mdivi-1 treatment decreases the expression of stemness markers SOX2, OCT4, and NANOG in DAZAP1-OE cells. Quantitative data are presented as the mean ± SD from at least 3 independent experiments. Statistical analysis by 1-way ANOVA followed by Tukey’s HSD post hoc test for multiple comparisons (B, G, H, and J) or unpaired Student’s t test (EV vs. OE) for comparisons between 2 groups and 1-way ANOVA followed by Tukey’s HSD post hoc test for multiple comparisons (shNC vs. sh1 and sh2) (F). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 7. DAZAP1 regulates alternative splicing of ULK1 to activate mitophagy.

(A) PCR array results demonstrated a marked reduction in ULK1 expression in shDAZAP1 cells. (B and C) qPCR and Western blot analyses indicated that DAZAP1 overexpression upregulated ULK1 mRNA and protein levels, while knockdown downregulated them. (D) RIP-qPCR analysis confirmed DAZAP1 binding to ULK1 RNA. (E) FISH-IF staining indicated colocalization of DAZAP1 and ULK1 RNA. Scale bars: 10 μm. (F) GSEA demonstrated substantial enrichment of the spliceosome pathway in DAZAP1-OE cells. (G) RNA-seq analysis identified alternative splicing sites within ULK1 RNA, showing that DAZAP1 promotes exon 17 skipping. (H) PCR and agarose gel electrophoresis confirmed DAZAP1 regulation of ULK1 RNA splicing. (I) Actinomycin D assays demonstrated that DAZAP1 increases ULK1 mRNA stability by reducing the production of premature termination codons and subsequent nonsense-mediated decay (NMD). (J) Co-IP experiments confirmed that DAZAP1 binds to several alternative splicing factors, including HNRNPC, DDX39B, HNRNPA1L2, HNRNPM, HNRNPA1, EIF4A3, PCBP1, RBMX, and HNRNPA3. (K) Co-IP confirmed the interaction between DAZAP1 and HNRNPA1, reinforcing DAZAP1’s role in alternative splicing regulation. (L and M) IHC analysis revealed a positive correlation between DAZAP1 and ULK1 expression levels (r = 0.69, P = 1.9 × 10^–5), n = 32. Scale bars: 200 μm. (N) Western blot analysis indicated that DAZAP1 overexpression upregulated mitophagy markers LC3B and P62, while knockdown downregulated them. Quantitative data are shown as the mean ± SD from a minimum of 3 independent experiments. Statistical analysis by unpaired Student’s t test (EV vs. OE) for comparisons between 2 groups and 1-way ANOVA followed by Dunnett’s multiple-comparison test (shNC vs. sh1 and sh2) (C), 1-way ANOVA followed by Dunnett’s multiple-comparison test (D), 2-way ANOVA followed by Šidák’s multiple-comparison test (I), or Spearman’s correlation analysis (M). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. NS indicates no statistically significant difference.

Figure 8. ULK1-mediated mitophagy is essential for DAZAP1-induced stemness in GC cells.

(A) Western blot analysis showed that ULK1 protein levels markedly decrease following DAZAP1 knockdown, while reintroducing ULK1 restores its expression. (B) Western blot analysis demonstrated that ULK1 overexpression rescues the expression of autophagy markers LC3B and P62 in DAZAP1-knockdown cells. (C and D) Mito-Keima labeling indicated that ULK1 overexpression partially restores mitophagy in DAZAP1-knockdown cells. (E) CCK8 assay showed that ULK1 overexpression partially restores cell proliferation in DAZAP1-knockdown cells. (F and G) Transwell migration assay indicated that ULK1 overexpression rescues the cell migration impaired by DAZAP1 knockdown. Scale bars: 100 μm. (H and I) Sphere formation assay demonstrated that ULK1 overexpression increases the number of spheres, counteracting the inhibitory effect of DAZAP1 knockdown. (J and K) qPCR and Western blot analyses showed that restoring ULK1 in DAZAP1-knockdown cells rescues the expression of stemness markers SOX2, OCT4, and NANOG. (L) ATP production assay indicated that ULK1 overexpression partially restores OXPHOS activity in DAZAP1-knockdown cells. (M) Gene expression analysis showed that ULK1 overexpression increases the expression of key OXPHOS complex subunit genes (UQCRC1, UQCRC2, SDHA, SDHB, ATP5F1A, ATP5F1B, and NDUFA1), indicating enhanced OXPHOS in DAZAP1-knockdown cells. Quantitative data are shown as the mean ± SD from a minimum of 3 independent experiments. Statistical analysis by 1-way ANOVA followed by Dunnett’s multiple-comparison test (D, G, I, J, L, and M) or 2-way ANOVA followed by Tukey’s HSD post hoc test for multiple comparisons (E). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. NS indicates no statistically significant difference.