DLGAP5 enhances bladder cancer chemoresistance by regulating glycolysis through MYC stabilization

Abstract

Rationale: Bladder cancer (BLCA), one of the most lethal urological tumors, exhibits high rates of recurrence and chemoresistance, particularly to gemcitabine (GEM). Understanding the mechanisms of GEM resistance is crucial for improving therapeutic outcomes. Our study investigates the role of DLGAP5 in promoting GEM resistance through modulation of glycolysis and MYC protein stability in BLCA cells. Methods: We utilized various BLCA cell lines and clinical tissue samples to analyze the impact of DLGAP5 on GEM resistance. Through biochemical assays, protein interaction studies, and gene expression analyses, we examined how DLGAP5 interacts with USP11 and MYC, assessed the effects on MYC deubiquitination and stability. The influence of these interactions on glycolytic activity and GEM resistance was also evaluated via mouse subcutaneous xenograft model and spontaneous BLCA model. Results: Our findings indicate that DLGAP5 enhances GEM resistance by stabilizing MYC protein via deubiquitination, a process mediated by USP11. DLGAP5 was found to facilitate the interaction between USP11 and MYC, promoting MYC-driven transcription of DLGAP5 itself, thereby creating a positive feedback loop. This loop leads to sustained MYC accumulation and increased glycolytic activity, contributing to GEM resistance in BLCA cells. Conclusion: The study highlights the critical role of DLGAP5 in regulating MYC protein stability and suggests that disrupting the DLGAP5-USP11-MYC axis may provide a novel therapeutic approach to overcome GEM resistance in BLCA. DLGAP5 represents a potential target for therapeutic intervention aimed at mitigating chemoresistance in bladder cancer BLCA.

Keywords: DLGAP5; MYC; bladder cancer; chemoresistance; glycolysis.

Figure 1

DLGAP5 enhances GEM resistance in BLCA cells. (A) Representative images (left panel) and statistical values (right panel) of IHC staining analysis of DLGAP5 protein levels in patients with BLCA (ZNWH cohort) treated with GEM chemotherapy (n = 24, ZNWH cohort_BCLA subgroup, details in Table S2). (B) Overall survival analysis of chemotherapy patients with different DLGAP5 mRNA levels in the TCGA BLCA dataset (n = 412). The optimal cut-point of the DLGAP5 mRNA expression was cut-off value. Patients with missing survival data were not included. The statistical significance of the survival data was ascertained by the log-rank test of Kaplan-Meier analysis. (C) Uniform Manifold Approximation and Projection (UMAP) visualization showing the expression levels of DLGAP5 in the epithelial cells of chemosensitive and chemoresistant tumors from a MIBC patient (GSE192575). (D) Cell viability of T24 cells with DLGAP5 knockdown after 48 h of treatment with various concentrations of GEM, as measured by MTT assay (n = 6). (E) Schematic overview showing the establishment of GEM-resistant cell lines (T24-R, UM-UC-3-R). (F) MTT assay results showing the viability of T24-R cells with DLGAP5 knockdown after 48 h of exposure to different concentrations of GEM (n = 6). (G) Statistical analysis of apoptosis in T24-R cells with DLGAP5 knockdown after 48 h of 10 μM GEM treatment (n = 3). (H) Western blot analysis of DLGAP5 proteins in UM-UC-3-P and UM-UC-3-R cells. (I) In vivo model construction and drug treatment (top). General view of dissected tumors from each group (bottom). (J) Weights of the tumors in each group (n = 5) after the tumors were surgically dissected. (K) Tumor growth of the indicated grafted mice treated with GEM was measured (n = 5). (L) Representative H&E (Scale bar = 100 μm) and IHC (Scale bar = 50 μm) staining analysis of subcutaneous tumor tissues from the xenograft models. Statistical significance of data was ascertained by two-tailed paired Student's t-test (A), two-tailed unpaired Student's t-test (J, K), and one-way ANOVA with Tukey's multiple comparisons test analyses (D, F, G). All statistical data are presented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 2

DLGAP5 influences GEM resistance in BLCA by regulating glycolysis. (A) Schematic of the identification of GEM resistance-related genes. Following dataset intersection, GSEA enrichment analysis revealed significant enrichment of the glycolysis signaling pathway. (B) Changes in the transcription levels of glycolysis-related genes in T24 cells after DLGAP5 knockdown (n = 3). (C) Western blot analysis of the effects of DLGAP5 knockdown on ENO1 and LDHA protein expression in T24 cells. Relative glucose uptake (D) and intracellular lactate production (E) in T24 cells after DLGAP5 knockdown (n = 3). Subcutaneous tumor models of BLCA were established in BALB/c nude mice via the injection of T24 control cells or T24 cells with DLGAP5 knockdown. Glucose uptake was analyzed via ¹⁸F-FDG PET-CT imaging (n = 4). Statistical values of SUVmax (F) and representative PET-CT imaging (G). Scale bar = 1 cm. (H) mRNA levels of glycolysis-related genes in T24-P and T24-R cells (n = 3). Relative glucose uptake (I) and intracellular lactate production (J) in T24-P and T24-R cells (n = 3). (K) T24-P and T24-R cells were treated with the indicated combinations of GEM (10 μM), 2-DG (2 mM), and oxamate (10 mM) before cell viability was measured at 48 h (n = 6). (L) siNC and siDLGAP5 T24 cells were treated with the indicated combinations of GEM (1 μM), pyruvate (2 mM), and lactate (10 mM) before cell viability was measured at 48 h (n = 6). Statistical significance of data was ascertained by two-tailed unpaired Student's t-test (F, H-L) and one-way ANOVA with Tukey's multiple comparisons test analyses (B, D, E). All statistical data are presented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3

The role of MYC in DLGAP5-mediated GEM resistance. (A) Hallmark gene sets (https://www.gsea-msigdb.org/gsea/msigdb) related to DLGAP5 from enrichment analysis of the gene expression matrix from RNA-seq assays. GSEA was performed with the R package “clusterProfiler”. p-value was computed via two-tailed Fisher's exact test. The Benjamini-Hochberg method was used to adjust the p-value. The top 15 gene sets were selected on the basis of the lowest p.adjust values and sorted in ascending order on the basis of the normalized enrichment score (NES), from largest to smallest. (B) GSEA of DLGAP5 knockdown in the hallmark MYC TARGET V1 gene set. (C) Western blot analysis of the effects of DLGAP5 knockdown on MYC proteins in T24 cells. (D) T24 cells were transfected with siDLGAP5 for 24 h, transfected with 5× E-box luciferase reporter for 48 h, and finally subjected to a dual-luciferase reporter assay (n = 3). (E) Western blots showing the protein expression of DLGAP5, HA-MYC, and LDHA in T24 cells after DLGAP5 knockdown and MYC overexpression. (F) Viability of T24 cells treated with various concentrations of GEM for 48 h, as determined via the MTT assay (n = 6). The asterisk indicates statistical significance between siD+Vector and siD+MYC. (G) Subcutaneous tumor models of BLCA were established in BALB/c nude mice via the injection of T24 shNC cells, T24 shDLGAP5 cells and T24 shDLGAP5+MYC cells. Glucose uptake was analyzed via ¹⁸F-FDG PET-CT imaging (n = 6). Scale bar = 1 cm. (H) General view of dissected tumors of each group (n = 3). (I) Weights of the tumors in each group after the tumors were surgically dissected (n = 3). (J) Representative IHC staining analysis of subcutaneous tumor tissues from xenograft models. Scale bar = 100 μm. (K) Representative MRI images of axial T2, showing the subcutaneous tumor xenografts. Scale bar = 1 cm. Statistical significance of data was ascertained by one-way ANOVA with Tukey's multiple comparisons test analyses (D, F, I). All statistical data are presented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4

DLGAP5 deubiquitinates and stabilizes MYC. After DLGAP5 was knocked down (A) or overexpressed (B) in UM-UC-3 cells, the mRNA level was detected via qRT-PCR (n = 3). (C) Western blot analysis of the effect of DLGAP5 knockdown on MYC degradation in T24 cells incubated with CHX (50 μg/mL) for the indicated time points. (D) UM-UC-3 cells were transfected with siNC or siDLGAP5 for 48 h and then treated with DMSO or MG132 (10 μM) or CQ (100 μM) for 8 h before lysis. Protein levels were analyzed by Western blotting. (E) Confocal imaging confirming that DLGAP5 co-localized with MYC in the nucleus of T24 cells. (F) Co-IP assay showing that exogenous DLGAP5 interacts with MYC in 293T cells. (G) 293T cells were transfected with GFP-DLGAP5 for 48 h, and a GST pull-down assay revealed that DLGAP5 interacts with MYC in vitro. Scale bar = 25 μm. (H) Schematic representation of various DLGAP5 truncations. (I) Co-IP assay showing that DLGAP5-NT interacts with MYC in 293T cells. (J) 293T cells were transfected with the specified plasmids for 48 h, followed by an 8 h of treatment with 10 μM MG132. Western blots showing exogenous ubiquitination of MYC after DLGAP5 knockdown in 293T cells. Statistical significance of data was ascertained by two-tailed unpaired Student's t-test (B) and one-way ANOVA with Tukey's multiple comparisons test analyses (A). All statistical data are presented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

The deubiquitinating enzyme USP11 regulates MYC stability and promotes GEM resistance. (A) Flow diagram showing the IP of DLGAP5 and subsequent LC-MS/MS analysis. (B) Co-IP assay showing that endogenous DLGAP5 and USP11 interact with MYC in UM-UC-3 cells. (C) Confocal imaging confirming that USP11 co-localized with MYC in the nucleus of T24 cells. The scale bar is 25 μm. (D) T24 cells were transfected with the specified plasmids for 48 h, followed by an 8 h of treatment with 10 μM MG132. Western blots showing exogenous ubiquitination of MYC after USP11 knockdown in 293T cells. (E) 293T cells were transfected with the specified plasmids for 48 h, followed by an 8 h of treatment with 10 μM MG132. Ubiquitination assays were conducted to examine the specific ubiquitin chain linkages catalyzed by USP11 on MYC proteins. (F) 293T cells were transfected with the specified plasmids for 48 h, followed by an 8 h of treatment with 10 μM MG132. Western blots showing exogenous ubiquitination of MYC after USP11 (WT) and USP11 (C318A) were overexpressed in 293T cells. (G) Co-IP assays showing that USP11 binds to MYC, particularly the N-terminal domain in 293T cells. (H) Schematic representation of various MYC deletion mutations (top) and co-IP assays showing that USP11 interacts with the MYC-NT MB1 domain in 293T cells (bottom). (I) 293T cells were transfected with the specified plasmids for 48 h, followed by an 8 h of treatment with 10 μM MG132. Ubiquitination experiments were conducted to identify the specific lysine residues on MYC that are deubiquitinated by USP11.

Figure 6

The DLGAP5-USP11-MYC feedback loop induces GEM resistance in BLCA cells. (A) 293T cells were transfected with the specified plasmids for 48 h, followed by treatment with 10 μM MG132 for 8 h. Western blots showed changes in exogenous ubiquitination of MYC. (B) Co-IP assays demonstrated that exogenous USP11 and MYC interactions decreased with DLGAP5 knockdown. (C) Western blot analysis of MYC proteins in T24-P and T24-R cells. (D) Co-IP assay showing that endogenous DLGAP5 and USP11 interact with MYC in UM-UC-3-P and UM-UC-3-R cells. (E) UM-UC-3-P and UM-UC-3-R cells were transfected with the HA-Ub for 48 h, followed by an 8 h of treatment with 10 μM MG132. Ubiquitination experiments evaluating the ubiquitination levels of MYC. (F) Spearman correlation analysis of the expression levels of DLGAP5 and MYC in the GEPIA database. (G) Representative images of IHC staining of DLGAP5 and MYC in human BLCA specimens from a BLCA tissue microarray (HBlaU050CS01). Scale bars, 100 μm. Pearson correlation analysis was used to determine the degree of association between DLGAP5 and MYC via IHC staining (n = 40). p-value was obtained by Student's t-test. (H) Genome browser tracks of MYC occupancy at the DLGAP5 locus in SKNAS and KELLY cells (GSE138295). The genome browser map is displayed via IVG software. The green region marks a region in the DLGAP5 promoter region where MYC is significantly enriched relative to the input. T24 cells with MYC knocked down (I) or MYC overexpressed (J), and the mRNA levels were detected by qRT-PCR (n = 3). (K) Dual-luciferase reporter assay of DLGAP5 promoter activity after overexpressing MYC in T24 cells (n = 3). (L) Binding site of MYC on promoter sequences was obtained from the JASPAR database. (M) Schematic diagram of primers designed for ChIP-qPCR of the DLGAP5 promoter sequence. (N) ChIP-qPCR analysis showed the enrichment degree of MYC in different regions of the DLGAP5 promoter. IgG indicates the negative control (n = 3). Statistical significance of data was ascertained by two-tailed unpaired Student's t-test (J, K, N) and one-way ANOVA with Tukey's multiple comparisons test analyses (I). All statistical data are presented as mean ± SD, * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 7

Mechanism diagram of the study. DLGAP5 drives BLCA GEM chemoresistance by facilitating glycolysis. In BLCA cells with high DLGAP5 levels, DLGAP5 facilitates MYC protein stability via the deubiquitinating enzyme USP11, which contributes to glycolysis. Enhanced glycolysis promotes an increase in metabolites such as lactate, leading to GEM chemoresistance. Additionally, MYC enhances DLGAP5 transcription by binding to its promoter region, forming a DLGAP5/USP11-MYC feedback loop that promotes GEM chemoresistance.