by Elizabeth Pennisi – Science Magazine
Our 21,000 protein-coding genes aren’t the only readable units in our genome. At last count, another 13,000 “genes” specify mysterious molecules called long noncoding RNAs (lncRNAs), and when the final tallies are in, they may outnumber protein-coding genes. But what are these RNAs good for? Some researchers have suggested that they represent “noise”: DNA randomly converted to RNA that serves no purpose. Others propose that they may be as pivotal as proteins in guiding cellular processes. To find out, Jesse Engreitz, a graduate student working with Mitchell Guttman and Eric Lander at the Broad Institute in Cambridge, Massachusetts, has taken a close look at one of the first noncoding RNAs discovered, XIST, which was identified 20 years ago as a silencer that shuts down one of the X chromosomes in females to ensure the proper amount of gene activity.
Engreitz has found that XIST operates by interacting with loops of nearby chromosome. “It seems to be creating a three-dimensional organization, bringing together regions of the genome in a way that we had assumed proteins were doing,” says Emmanouil Dermitzakis, a genomicist from the University of Geneva in Switzerland. This finding supports a role for lncRNAs in regulating chromosomal activity by influencing the shape of chromatin, the protein complex that swaddles DNA. “It gives us a model of how other lncRNAs might be active,” Dermitzakis adds.
Discovered in the early 1990s, XIST—along with the few other long noncoding RNAs known at the time—was considered an anomaly. XIST’s gene is located on the X chromosome. As it converts to RNA, XIST spreads over the X chromosome, silencing genes. After 2 decades of study, researchers still do not know how this spreading occurs or how XIST recognizes which parts of the X to inactivate.
When Engreitz arrived in Guttman’s lab 2 years ago, the team was developing a way to see where along the genome a particular lncRNA would bind. Together, they came up with a method that uses RNA probes complementary to the lncRNA to target, bind, and precipitate out parts of the genome. When Engreitz tested this approach with XIST, he found that it bound to the X chromosome, but not where he expected. “It seems to bind everywhere,” he said.
The scientists wondered if chromatin’s 3D arrangement might come into play. Other researchers had used a method called Hi-C to build a 3D map of the twists and turns of the X chromosome. When Engreitz and his colleagues compared this map to their map of where XIST begins to bind, they saw a tight correlation with twists and turns close to where the XIST gene was located. “Where XIST goes first are the [DNA] sites that contact the XIST [gene],” he reported at the meeting.
In one experiment, Engreitz and his colleagues moved XIST 50 million bases down the X chromosome and put that altered X chromosome in mice embryonic stem cells. XIST interacted with a new set of DNA loops nearby. And when they put the XIST gene on a different chromosome, they saw a similar shift in binding. The results “clearly showed that physical proximity and interaction with the chromatin, and not sequence specificity, is important for spreading X-inactivation,” says Piero Carninci from the RIKEN Center for Life Science Technologies in Yokohama, Japan. “This is quite impressive.”
Other studies have shown that as XIST inactivation proceeds, XIST seems to reel in the outer loops of the X chromosome, possibly by recruiting proteins that alter chromatin’s conformation. “It’s possible that lncRNAs represent a new type of gene regulator,” says Rory Johnson, a genomicist at the Centre for Genomic Regulation in Barcelona, Spain.
Preliminary results with other lncRNAs suggest that they, too, may work like XIST, Engreitz reported. Other researchers point out that lncRNAs are abundant and may work in many different ways. “We just don’t know,” Johnson says.