Enhancer Journal Club: eRNAs are required for p53-dependent enhancer activity and gene transcription.
Back to eRNAs, and this time we're going to look at possible roles in DNA damage control. This paper by Melo et al. is from Reuven Agami's lab at the Netherlands Cancer Institute. The abstract is as follows:
Binding within or nearby target genes involved in cell proliferation and survival enables the p53 tumor suppressor gene to regulate their transcription and cell-cycle progression. Using genome-wide chromatin-binding profiles, we describe binding of p53 also to regions located distantly from any known p53 target gene. Interestingly, many of these regions possess conserved p53-binding sites and all known hallmarks of enhancer regions. We demonstrate that these p53-bound enhancer regions (p53BERs) indeed contain enhancer activity and interact intrachromosomally with multiple neighboring genes to convey long-distance p53-dependent transcription regulation. Furthermore, p53BERs produce, in a p53-dependent manner, enhancer RNAs (eRNAs) that are required for efficient transcriptional enhancement of interacting target genes and induction of a p53-dependent cell-cycle arrest. Thus, our results ascribe transcription enhancement activity to p53 with the capacity to regulate multiple genes from a single genomic binding site. Moreover, eRNA production from p53BERs is required for efficient p53 transcription enhancement.
As always, I've posted a non-technical summary over at ScienceGist: http://sciencegist.com/gists/209
This is a nice paper that fits with most of the other eRNA papers published in similar areas this year. Like others, they start by investigating the binding of a particular transcription factor (in this case P53) outside of genes, and find that this extra-genic binding correlates with known markers of enhancers (in this case H3K4me1, K3K27Ac and P300). They test these regions in a luciferase assay and find that they do indeed show enhancer activity, and crucially they show that this enhancer activity is at least partially P53 dependent by simultaneously knocking down P53.
They want to identify possible target genes, and they decide to do this by identifying enhancer/promoter loops using a 4C screen. Using a chemical called nutlin which increases stability (therefore steady state concentration) of P53, they show that increased P53 levels increase expression of some identified target genes. Again, as a nice control they also do a simultaneous knock down of P53 to show that the effect of the nutlin treatment is really P53 dependent. Not only is the transcription of nearby genes activated by the presence of P53, but also the production of enhancer RNAs from the enhancer loci themselves. This leads to the most interesting experiment, where they try to investigate whether it is the eRNA or the enhancer DNA which is responsible for activation. They do this by creating a chimeric transcript where the eRNA is fused to 24 copies of MS2 RNA. They can then tether this RNA to a luciferase reporter using an MS2-CP:GAL4 fusion protein. This tethering does indeed seem to increase the transcription of the luciferase, and activation is not seen when a mutated transcript is used. This is a really nice result, but much of the activation effect is observable even when no MS2-CP:GAL4 fusion protein is added. This appears to indicate that the presence of the eRNA in the nucleus activates the expression of the luciferase even when it is not physically present at the luciferase promoter, which I personally can't make sense of. The authors merely say that "The remaining MS2-CP-independent transcriptional activation can be attributed to the overexpression of the eRNA-MS2 in the nucleus." - so perhaps I am missing something very obvious here...
Finally, they tie the eRNA producing regions back to a functional effect. One of the primary functions of P53 is to arrest the cell cycle following DNA damage, so they expose the cell to ionizing radiation whilst simultaneously knocking down the eRNA. This experiment shows that knock down of the P53 induced eRNA allows a partial bypass of the normal cell cycle arrest, although of course the effect is not as dramatic as knocking down P53 itself.
All in all, this is a very interesting paper that provided one of the first insights into a functional role for eRNAs. My feeling at the moment is that the slew of papers about eRNAs this year have demonstrated an important function for eRNAs but even taken together, the evidence for a cis-acting mechanism as opposed to eRNAs being another class of trans-acting ncRNAs is not yet air-tight.