Shining a Brighter Light on Genomic Dark Material

Abstract

“Perhaps the biggest surprise of the post-genomic era has been the enormous number and diversity of transcriptional products arising from the previously presumed wastelands of the non-protein-coding genome,” write John Mattick and John Rinn of the Garvan Institute of Medical Research and Harvard University, respectively, in the January 2015 issue of Nature Structural and Molecular Biology. “In fact,” David Spector of the Cold Spring Harbor Laboratory points out, “70 to 80 percent of the human genome is likely to be transcribed, yielding a complex network of overlapping transcripts that includes tens of thousands of long ribonucleic acids (RNAs) with little or no protein-coding capacity.” Many long non-protein-coding RNAs (lncRNAs) are proving to be key regulators of gene expression, whereas others act as molecular decoys, serve as small RNA precursors, act as scaffolds for protein–protein interactions, or drive nuclear organization. In addition, “emerging evidence suggests that a number of lncRNAs help guide epigenetic complexes to their sites of action, control chromatin architecture, and participate in embryo development,” says Mattick. To complicate matters even further, some lncRNAs do have, in addition to their primary talents, the ability to encode proteins—albeit usually only small ones. Ron Breaker and his colleagues at Yale are making significant progress identifying the roles of several giant, very complicated noncoding RNAs discovered in extremophilic bacteria, whose mysterious functions sometimes help the cells cope with unusual environmental stresses. An ornate, large extremophilic (OLE) RNA recently identified in Bacillus halodurans, for example, seems to protect its host bacterium against both alcohol-induced stress and cold. However, two other newly discovered structures—GOLLD (giant, ornate, lakeand lactobacillales-derived) and HEARO (HNH endonucleaseassociated RNA and open reading frame)—so far appear to do bacteria more harm than good. Notably though, both GOLLD and HEARO have the distinction of being among the most complex, giant RNAs ever encountered. GOLLD is “particularly striking because it’s the third biggest highly structured bacteria-associated RNA ever described, ranking only behind 23S and 16S ribosomal RNAs,” according to the Breaker Lab’s Zasha Weinberg. But GOLLD sits on a prophage, probably helping it to survive, mature into an adult virus, and infect; therefore, it is no friend to bacteria. Similarly, HEARO looks and acts like a mobile genetic element and replicates promiscuously in its target bacteria, which could do some harm, at least in the short term. Eventually, however, the challenge to its hosts might drive evolution, which could make HEAROinfected bacteria better able to deal with their environments. (The benefits of genetic conflict are detailed in “Rules of conduct: Ancient battles between viruses and hosts,” in the June 2014 issue of BioScience (http://io.aibs. org/virhost). In contrast to bacterial lncRNAs, the functions of most lncRNAs so far identified in multicellular organisms are still unknown. However, many more of these intriguing regulatory RNAs in both simple and complex organisms await discovery, say the experts. Therefore, identifying new lncRNAs and figuring out what they do has grown into a robust scientific industry, keeping hundreds or maybe even thousands of researchers around the world busy. The recent explosion of new technologies is expected to make their jobs easier. Long noncoding RNAs, for example, can now be identified with high-throughput sequencing instead of optical imaging, says Stanford University’s Howard Y. Chang, who considers this is a major advance. “Likewise, traditional in vitro purification methods are being replaced by [more accurate] in vivo testing,” adds Ci Chu, also at Stanford. Furthermore, low-throughput RNA-mapping has been combined with next-generation sequencing, and that helps provide a wider range of information. Notably, lncRNA itself is a vague catch-all grouping into which any transcript that doesn’t appear to have protein coding as its primary job and is larger than 200 nucleotides in length qualifies for inclusion. “The size is arbitrary but has the advantage of being a convenient biophysical cutoff that excludes most known, albeit also poorly understood, classes of small regulatory and other RNAs. However, even messenger RNAs sometimes moonlight as functional lncRNA,” Chang says. But even though alternative terminology has been suggested, for the time being, RNAists have settled comfortably (or not) on lncRNA. “As confusing as it might sometimes be, the complexity and interconnectedness of lncRNAs should not be an impediment to but rather the motivation for exploring the vast unknown universe of RNA regulation, without which we will not understand biology,” argues Mattick.