Identification and functional characterization of long non-coding RNAs associated with epithelial-to-mesenchymal transition

In the last decade, new high-throughput sequencing techniques have revealed the complexity of the human transcriptome, allowing the characterization of long non-coding (lnc)RNAs. Since their expression has been reported as very specific to tissue, developmental stage and pathological variations, some lncRNAs have been proposed as biomarkers for diagnosis as well as prognosis of tumors. In this project, we aime in identification of lncRNAs associated with the epithelial-to-mesenchymal transition (EMT), a biological process that promotes metastasis. This was performed on an in vitro model (Castro Vega et al, 2015) in which primary HEK cells naturally undergo EMT. Cells were immortalized prior and after EMT, therefore giving rise to the Epi (HA5-Early) and Mes (HA5-Late) cell-lines, which respectively display epithelial and mesenchymal phenotypes. HOTAIR is one the lcnRNAs functionning as a scaffold to tether PRC2 and Lsd1/REST/coREST complexes to gene promoter to epigenetically repress transcription. Its high expression is associated with poor-prognosis cancer and promotes EMT. To decipher a role of PRC2 and Lsd1 interacting domains in HOTAIR function we generated Epi cell lines expressing full length and truncated versions of HOTAIR missing first 5'-300 bp and last 3'-500 bp interacting with HOTAIR partners. Total RNA form these cell lines was extracted and sequenced to reveal HOTAIR targets dependent of its intercation with PRC2 and/or Lsd1. The experiment was performed in biological duplicates of each cell-line (Epi-CTR, Epi-HOT, Epi-HOTΔP and Epi-HOTΔL).