Comprehensive transcriptome and epigenome sequencing of hypoxic breast cancer reveals non-coding RNAs associated with clinicopathological features. H. Choudhry1,2, J. Schodel3, S. Oikonomopoulos1, C. Camps1, P. Ratcliffe3, A. Harris4, D. Mole3, I. Ragoussis1 1) Nuffield Department of Clinical Medicine, Wellcome Trust Centre of Human Genetics, University of Oxford, Oxford, United Kingdom; 2) Biochemistry Department, Faculty of Science, King Abdulaziz University, Jeddah, Saudi Arabia; 3) Henry Wellcome Building for Molecular Physiology, University of Oxford, Oxford, United Kingdom; 4) Department of Oncology, John Radcliffe Hospital, University of Oxford, Oxford, United Kingdom.
Hypoxia is an important physiological stress with a central role in many responses including adaptation to altitude, growth and development, as well as major pathophysiological processes such as ischaemic and cancer. Pan-genomic analyses of these responses have focussed on protein-coding genes. However, it is now recognised that many transcripts have functions that do not include coding for proteins (non-coding RNA). Here, we comprehensively profile this non-coding transcriptional output in hypoxia. We undertook an integrated pan-genomic analysis of normoxic and hypoxic MCF7 breast cancer cells, employing RNA-seq together with ChIP-seq for the major hypoxia-inducible transcription factor HIF and for chromosomal markers of active transcription (RNApol2 and histone H3K4 methylation). We further assessed these responses following knockdown of HIF transcription factors using RNAi. We establish a computational pipeline of strand specific ribo-depleted RNA-seq data to detect regulated non-coding transcripts including piwiRNA, miRNA, tRNA, sn/snoRNA, and lncRNA. Compared to other classes snRNAs and tRNAs are globally downregulated, whilst a significant number of lncRNAs are up-regulated. These up-regulated lncRNAs are associated both with chromosomal markers of transcriptional activation and with HIF binding, indicating direct transcriptional activation of non-coding transcripts by HIF. Dependence of lncRNAs on HIF was further confirmed by HIF RNAi. In addition we describe 105 novel previously unannotated transcripts bearing chromosomal marks of bon-fide genes. Four hypoxically induced lncRNAs were then analysed in 148 breast tumours and associated with clinicopathological features.Our findings extend knowledge of the hypoxic transcriptional response into the spectrum of non-coding transcripts. We demonstrate that HIF can transcriptionally activate lncRNAs in addition to coding transcripts and link these to clinicopathological features. These HIF-regulated non-coding transcripts have the potential to act as new biomarkers for breast cancer as well as potentially novel therapeutic targets.
You may contact the first author (during and after the meeting) at