Analysis of open chromatin regions in bladder cancer links β-catenin mutations and Wnt signaling with neuronal subtype of bladder cancer.

Urothelial carcinoma of the bladder is the most frequent bladder cancer affecting more than 400,000 people each year. Histopathologically, it is mainly characterized as muscle invasive bladder cancer (MIBC) and non-muscle invasive bladder cancer (NMIBC). Recently, the studies largely driven by consortiums such as TCGA identified the mutational landscape of both MIBC and NMIBC and determined the molecular subtypes of bladder cancer. Because of the exceptionally high rate of mutations in chromatin proteins, bladder cancer is thought to be a disease of chromatin, pointing out to the importance of studying epigenetic deregulation and the regulatory landscape of this cancer. In this study, we have analyzed ATAC-seq data generated for MIBC and integrated our findings with gene expression and DNA methylation data to identify subgroup specific regulatory patterns for MIBC. Our computational analysis revealed three MIBC regulatory clusters, which we named as neuronal, non-neuronal and luminal outlier. We have identified target genes of neuronal regulatory elements to be involved in WNT signaling, while target genes of non-neuronal and luminal outlier regulatory regions were enriched in epithelial differentiation and drug metabolism, respectively. Neuronal regulatory elements were determined to be ß-catenin targets (p value = 3.59e-08) consisting of genes involved in neurogenesis such as FGF9, and PROX1, and significantly enriched for TCF/LEF binding sites (p value = 1e-584). Our results showed upregulation of ß-catenin targets regulated by neuronal regulatory elements in three different cohorts, implicating ß-catenin signature in neuronal bladder cancer. Further, integration with mutation data revealed significantly higher oncogenic exon 3 ß-catenin mutations in neuronal bladder cancer compared to non-neuronal (odds ratio = 31.33, p value = 1.786e-05). Our results for the first time identify regulatory elements characterizing neuronal bladder cancer and links these neuronal regulatory elements with WNT signaling via mutations in β-catenin and its destruction complex components.