Construction of a Co-Expression Network for lncRNAs and mRNAs Related to Urothelial Carcinoma of the Bladder Progression

Abstract

Carcinoma of urinary bladder is the most familiar cancer of the urinary tract, with the highest incidence in men. However, its prognosis and treatment have not improved significantly in the last 30 years. The main reason for this may be related to the alteration and regulation of genes. These alterations in genes that play a crucial role in cell cycle regulation may result in high-grade tumors and may alter drug sensitivity. Notably, the role of lncRNA in bladder cancer, especially the lncRNA-mRNA regulatory network, has not been fully elucidated. In this manuscript, we compared RNA sequencing (RNA-seq) data from 19 normal bladder tissues and 411 primary bladder tumor tissues using The Cancer Genome Atlas (TCGA) data bank, subjected differentially expressed mRNAs and lncRNAs to weighted gene co-expression network analysis, and screened out modules highly correlated with tumor progression. Subsequently, a lncRNA-mRNA co-expression network was built, and two key mRNAs were identified via COX regression analysis. Kaplan-Meier curve analysis revealed that the overall survival of sick people in the high-risk section was significantly shorter than those in the low-risk section. Therefore, this lncRNA-mRNA-based co-expression pattern may be used clinically to predict the prognosis of carcinoma of urinary bladder people. Our study not only provides a genetic target for carcinoma of urinary bladder therapy but also provides new ideas for people in the medical profession to discover the treatment of various tumors.

Keywords: COX analysis; WGCNA; lncRNA-mRNA co-expression network; prognostic marker; urothelial carcinoma of the bladder.

Figure 1

Differentially expressed lncRNAs and mRNAs in TCGA-BLCA. [(A) volcano diagram showing BLCA differentially expressed lncRNA in TCGA; (B) volcano diagram showing BLCA differentially expressed mRNA in TCGA].

Figure 2

Construction of weighted gene co-expression network. [(A) selection of soft threshold β; (B) correlation heat map of neighboring modules in WGCNA; (C) clustering dendrogram of gene modules; (D) in the module feature heat map, each column corresponds to clinical parameters and each row corresponds to a module feature gene].

Figure 3

Enrichment analysis of genes in the blue module. [(A) Top 20 enrichment of KEGG pathway of the mRNA in a blue module; (B) Top 20 enrichment of GO pathway of the mRNA in the blue module; (C) Interaction and overlapping (of the top 10 pathways)].

Figure 4

lncRNA-mRNA co-expression network and PPI network construction. [(A) lncRNA-mRNA co-expression network in the blue module; (B) protein-protein interaction network of co-expressed mRNAs in co-expression network; (C) MCODE plug-in screening out the highest-scoring sub-network; (D) co-expression network of lncRNA DEPDC1-AS1 and mRNAs in MCODE1].

Figure 5

Expression and prognostic analysis of lncRNA and mRNA in BLCA. [(A) DEPDC1-AS1 expression in BLCA; (B) CCNB1 expression in BLCA; (C) CDC20 expression in BLCA; (D) correlation between DEPDC1-AS1 and CCNB1 expression in BLCA; (E) correlation between DEPDC1-AS1 and CDC20 expression in BLCA; (F) KM curve of DEPDC1- AS1 expression with KM curve and time-dependent ROC curve of overall survival; (G) KM curve and time-dependent ROC curve of CCNB1 expression with overall survival; (H) KM curve and time-dependent ROC curve of CCNB1 expression with overall survival].

Figure 6

Gene set enrichment analysis. [(A) REACTOME and HALLMARK pathway analysis of CCNB1; (B) BREACTOME and HALLMARK pathway analysis of CDC20].

Figure 7

Relative expression of genes quantified by qRT-PCR. [(A) Relative expression level of lncRNA DEPDC-AS1; (B) Relative expression level of CCNB1; (C) Relative expression level of CDC20; ***P < 0.001].