UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Endogenous retroviruses drive transcriptional innovation in human cancer Artem, Babaian

Abstract

Transposable element (TE) exaptation is the process of TE incorporation into functional, and in some cases necessary, genes or regulatory units over evolutionary time. I postulate that an analogous process occurs in oncogenesis, wherein TE-derived promoters generate “noisy” transcription and novel transcripts which can then undergo selection to drive cancer transcriptome evolution. Such “onco-exaptation” is reviewed in the context of several cancers including Hodgkin Lymphoma (HL) where it results in expression of the oncogene CSF1R, yet it is unclear how widespread this phenomenon is. I hypothesize that epigenomic dysregulation in cancer leads to a genome-wide derepression of TE-initiated transcripts, some of which have an oncogenic role. To address this hypothesis, I developed a computational tool called ‘LIONS’ to analyze RNA-sequencing data for TE-initiated transcripts. LIONS detects and quantifies TE-initiated transcripts through transcriptome assembly, applies a novel artificial neural network classifier to identify TE promoter events, and compares biological sets of data. Using this tool, I have determined that the transcriptomes of colorectal carcinoma, diffuse large B-cell lymphoma and HL all have an overall increase in TE-initiated transcripts relative to their respective controls. This increase is specifically driven by an increase in endogenous retroviral long terminal repeat (LTR) initiated transcripts. The distribution of this TE transcriptional activity is widely distributed across the genome, yet patterns of co-activation among element families and the recurrent activation of a small sub-set of TEs is evident. One such recurrent TE-initiated transcript is the LOR1a LTR driven expression of the IRF5 oncogene in HL. IRF5, along with CSF1R and a panel of putative oncogenic TE-initiated transcripts were explored as novel biomarkers in HL. Altogether, I propose that the process of onco-exaptation is a novel and distinct mechanism for oncogene activation and a model system for future studies of exaptation and transcriptome evolution.

Item Citations and Data

Rights

Attribution 4.0 International