Bioinformatics analysis and experimental approach identify BRD-family gene DSP as a diagnostic and prognostic...

As a highly prevalent malignant tumor within the urinary system, bladder cancer (BLCA) is manifested as frequent recurrence and substantial prognostic heterogeneity. The development of BLCA is strongly linked to epigenetic dysregulation. Bromodomain-containing proteins (BRDs)-related genes (BRDRGs) serve as critical regulators across various cancers. However, the expression profiles, prognostic significance, and influence on the tumor immune microenvironment (TIME) of BRDRGs in BLCA remain underexplored. This study aimed to systematically explore the expression and prognostic significance of BRD-related genes in BLCA and to construct a BRD-based prognostic model.

Utilizing transcriptomic data within The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) for BLCA, BRDRGs were screened through GeneCards. Consensus clustering was leveraged to define BRD molecular subtypes. Differential expression analysis was implemented to ascertain BRD co-expressed genes, and we intersected them with differentially expressed genes (DEGs) between tumor and normal samples in TCGA-BLCA to obtain differentially expressed BRD genes (DE-BRDRGs). A prognostic risk model was developed utilizing univariate Cox and least absolute shrinkage and selection operator (LASSO) regression, followed by survival analysis, receiver operating characteristic (ROC) evaluation, and nomogram validation. Associations between the model and immune characteristics with immunotherapy response were estimated utilizing Cell-type Identification By Estimating Relative Subsets Of RNA Transcripts (CIBERSORT), Tumor Immune Dysfunction and Exclusion (TIDE), and pRRophetic analyses. Gene set enrichment analysis (GSEA) was implemented to ascertain functions of key genes. In vitro, the key gene DSP was knocked down utilizing small interfering RNA (siRNA). Reverse transcription quantitative polymerase chain reaction (RT-qPCR) and Western blot were implemented to verify the effectiveness of knockdown, while Cell Counting Kit-8 (CCK-8) was leveraged to evaluate the proliferation of BLCA cells.

In total, 1,572 upregulated genes and 1,111 downregulated genes were ascertained from TCGA-BLCA, with 36 DE-BRDRGs ultimately ascertained through the intersection of BRD co-expressed genes. Seven genes (DSP, DSC2, HSPG2, GJA1, TXNRD1, PCDHGC3, and PEG10) were chosen to formulate the risk model. In TCGA-BLCA, the area under the curve (AUC) values within 1, 2, and 3 years were 0.620, 0.665, and 0.682. In GSE13507, the AUC values within 1, 2, and 3 years were 0.730, 0.655, and 0.664. Marked distinctions were observed between high- and low-risk cohorts regarding immune cell infiltration (e.g., Tregs and CD8⁺ T cells) and immunotherapy response. GSEA indicated that key genes contributed to the progression of BLCA through pathways like the cell cycle and PI3K-Akt signaling. In vitro experiments denoted that DSP knockdown considerably inhibited the proliferation of BLCA cells, and the knockdown was effective, suggesting its pro-tumorigenic role.

This research ascertained seven key BRDRGs and comprehensively analyzed their functions in BLCA. In vitro experimental validation demonstrated that DSP exerts tumor-promoting effects and represents a potential therapeutic target. The established BRD-based prognostic model offers a novel strategy for risk stratification and personalized management of individuals with BLCA.