Dr. John L. Rinn from Harvard University and the Broad Institute (Harvard and MIT) is a prominent researcher in long non-coding RNAs (lncRNAs) and agreed to sit down and discuss his perspective on the intersection of lncRNA and epigenetics.
EpiGenie: You’ve mentioned that you feel knockout models are going to play a key role in the next phase of lncRNA studies in your lab. Do you see this as the natural progression in the field of long non-coding RNA in general?
Dr. Rinn: I think the field is at the point where it’s growing up. It’s maturing and it’s beginning to set a higher bar. There have been a few mechanistic studies that examine these next steps, but I think the other thing that’s happened in the field is that these lncRNAs are so bizarre that they almost seem like science fiction. So what’s happening now is that because these conclusions are so out there, people expect a higher burden of proof to back some of these claims to balance the excitement.
I think the implications have all been there, but nobody’s actually seen it from mouse to mechanism. The only clear cut lncRNA on a genetic and molecular level we really know about is Xist, which is this big whopping cloud of RNA and chromatin shutting off nearly an entire chromosome; a power to behold.
So this idea of RNA as a scaffold is going to cause a paradigm shift. It could quickly change from the idea of this RNA binds to this protein and moves it to this location to more of a like a polymer scaffolding of nuclear architecture in three dimensions.
EpiGenie: So why do you think we’re not there yet? Is it a lack of technology or interest, or some skepticism?
Dr. Rinn: I think people don’t know where to start. It’s very intimidating when you profile and you see 8,000 different things. I think 90% of collaborations start out with us generating a very expensive data set, which contains a ton of lncRNAs that are changing in a system.
But then you have to commit three to four years to figure out what’s going on after that. So I think a lot of people have stopped at the point of survey because that’s the easier thing to do. And it’s not that researchers are lazy but rather that the tools require a lot of commitment and a lot of expertise that are now emerging.
However, now I think the time has arrived. The field is getting more and more standardized. Now when most researchers profile something, they do a loss of function screen, find out which transcripts have a phenotype, and then figure out the mechanism. I think that this will be the minimal barrier to publishing soon.
EpiGenie: So, is there any evidence that suggests that lncRNAs are behind structural changes in chromatin, or is it rather structural changes in chromatin that allow for lncRNA binding?
Dr. Rinn: I like to think of the nucleus as a discreet “structure” or “building”. Although buildings have different shapes they all need beams and scaffolds. We have been seeing that RNA might be more of a generic polymer to facilitate unique architecture. This would also explain why lncRNAs might be in low abundance, yet many may contain similar structural elements (like beams)..
I want to think the RNA is pulling together the nuclear architecture, but we can’t conclude that just yet. However, there was an experiment that showed that if you take away the nuclear matrix protein that binds RNA, the RNA scatters all the way out into the cytoplasm.
EpiGenie: We know that RNA can undergo a plethora of modifications. Do you think that there are there any discreet marks that help dock RNAs into the matrix?
Dr. Rinn: With the recent birth of the field of epi-transcriptomics, researchers have begun to examine marks like N6-methyl-A and pseudouridylation. However, I’ve stayed away from those things towards a personal goal of testing function through genetics. However that could also be considered short-sightedness on our part.
EpiGenie: It sounds like there is more of a need for a shift into not only functional studies, but also visualization. Are there tools out there today that really get at it and look past the mapping?
Dr. Rinn: So the next step is in vivo knockout phenotypes. This will allow for situations where you can see tissues disrupted or things that you can see with your eye and you know something is wrong. Indeed, the living cell is where we’re developing methods right now for tracking them as well.
There’s this interesting phenomena that lncRNAs are low abundance, but does that mean there is a low abundance in all the cells or is there a certain context or specific populations that are blasting out lncRNAs? We’re actually seeing some of both cases, where lncRNAs must actually be one copy per cell, and what that means is that we’re not totally sure yet. Then there’s the other case where it’s blasting out copies in one out of seven cells, but not present in any of the surrounding ones.
EpiGenie: So, if there was one hypothesis you or your lab could answer in the next six months, what would it be?
Dr. Rinn: Can RNA be mitotically inherited and adaptable through environment? That’s what I think the future of it is. If we think about the Holy Grail about molecular biology, it’s currently how does the cell split, replicate, erase all these marks, and then reestablish the same thing.
If you think about the sperm, it’s even more ridiculous where it wipes out the entire epigenome and then re-establishes it in two days. So there’s some premise of an active remodeling of inheritance of cell state.
I would love to know if there’s mitotically inherited lncRNAs that can help establish and/or re-establish the chromatin state when cells replicate. Another intriguing idea is RNA serving as an evolutionary buffer of sorts. Similar to the notion that kids play in more dirt challenge have better adapted immune systems. Perhaps the genome is getting dirty with transcription and being selected by or buffering insults from the environment. That is the exciting part of the field is the possibilities, but for now it’s back to the reality of trying to chain these genes down one link at a time.