Final Report Summary - POLYLOOP (Policing mammalian transcriptomes: regulation of long ncRNA synthesis by transcriptional termination, gene loops and R-loops.)
It is well established that RNA polymerase II (Pol II) synthesises protein encoding transcripts (pre-mRNA) that are processed into mRNA that then act as templates for protein synthesis. However a surprising feature of the mammalian transcriptome is that Pol II also makes a plethora of long non coding (lnc)RNA. It was initially assumed that lncRNA would be transcribed and processed in a similar manner to pre-mRNA. The principal objective of my ERC programme was to test whether lncRNA employ different expression strategies and how these might relate to functionality. Remarkably we have shown in a series of interconnected research projects (many already published in high profile papers) covered by the umbrella of this research programme that lncRNA employ radically different transcription and RNA processing strategies. Firstly Pol II generates a substantial fraction of lncRNA through the capacity of R-loop structures found near promoters, terminators and enhancers to act as de novo Pol II promoters. Subsequently these transcripts are processed and terminated by a range of different mechanisms. Thus the termination process of lncRNA appear to substantially differ from pre-mRNA. Also coupled RNA processing (splicing and 3' end cleavage and polyadenylation) does not occur efficiently on most lncRNA which instead are often co-transcriptionally degraded. Overall our studies funded by this ERC "polyloop" grant lead us to the view that many if not most lncRNA are generated by default and have no intrinsic sequence specific function. Instead we see that they are restricted in amount and extent by nuclear quality control mechanisms. Significantly lncRNA over-expression is highly detrimental to the cell and is likely to be an important feature of cancer cell progression. The minority of functional lncRNA must somehow escape these restrictive mechanisms.