Molecular subtypes of urothelial carcinoma are defined by specific gene regulatory systems

BMC Med Genomics. 2015 May 26:8:25. doi: 10.1186/s12920-015-0101-5.

Abstract

Background: Molecular stratification of bladder cancer has revealed gene signatures differentially expressed across tumor subtypes. While these signatures provide important insights into subtype biology, the transcriptional regulation that governs these signatures is not well characterized.

Methods: In this study, we use publically available ChIP-Seq data on regulatory factor binding in order to link transcription factors to gene signatures defining molecular subtypes of urothelial carcinoma.

Results: We identify PPARG and STAT3, as well as ADIRF, a novel regulator of fatty acid metabolism, as putative mediators of the SCC-like phenotype. We link the PLK1-FOXM1 axis to the rapidly proliferating Genomically Unstable and SCC-like subtypes and show that differentiation programs involving PPARG/RXRA, FOXA1/GATA3 and HOXA/HOXB are differentially expressed in UC molecular subtypes. We show that gene signatures and regulatory systems defined in urothelial carcinoma operate in breast cancer in a subtype specific manner, suggesting similarities at the gene regulatory level of these two tumor types.

Conclusions: At the gene regulatory level Urobasal, Genomically Unstable and SCC-like tumors represents three fundamentally different tumor types. Urobasal tumors maintain an apparent urothelial differentiation axis composed of PPARG/RXRA, FOXA1/GATA3 and anterior HOXA and HOXB genes. Genomically Unstable and SCC-like tumors differ from Urobasal tumors by a strong increase of proliferative activity through the PLK1-FOXM1 axis operating in both subtypes. However, whereas SCC-like tumors evade urothelial differentiation by a block in differentiation through strong downregulation of PPARG/RXRA, FOXA1/GATA3, our data indicates that Genomically Unstable tumors evade differentiation in a more dynamic manner.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Binding Sites
  • Biomarkers, Tumor / metabolism
  • Carcinoma / classification
  • Carcinoma / genetics*
  • Cell Line, Tumor
  • Chromatin Immunoprecipitation
  • Cluster Analysis
  • Computational Biology
  • Fatty Acids / chemistry
  • Gene Expression Profiling*
  • Gene Expression Regulation, Neoplastic*
  • Humans
  • Lipids / chemistry
  • Mice
  • Promoter Regions, Genetic
  • Protein Binding
  • Software
  • Transcription, Genetic
  • Urinary Bladder Neoplasms / classification
  • Urinary Bladder Neoplasms / genetics*
  • Urothelium / pathology*

Substances

  • Biomarkers, Tumor
  • Fatty Acids
  • Lipids