Science http://dx.doi.org/10.1126/science.aaf5371 (2016)

In vitro, some G-rich RNA sequences can fold into four-stranded structures called RNA G-quadruplexes (RG4s), which are stable in the presence of K+ ions. Although a number of techniques have hinted at the presence of RG4s in cells, it is unclear whether the folding observed in vitro reflects folding in cells. To test this, Guo and Bartel developed a method to measure the folding states of endogenous RNA transcripts at a global level by coupling methylation of guanine residues with the capacity of RG4s to stall reverse transcriptase, and then deep sequencing the resulting complementary DNA fragments. They verified in vitro that folded RG4 structures produced truncated transcripts, whereas unfolded RG4 structures produced longer transcripts. Surprisingly, analysis in mouse embryonic stem cells and yeast revealed that most of the RG4 regions were unfolded, despite the high intracellular of K+ in these cells, and RG4 constructs ectopically introduced into either cell type were similarly unfolded. In contrast, analysis of the Escherichia coli transcriptome revealed far fewer RG4 structures, suggesting that RG4-forming sequences were eliminated during bacterial evolution. Consistent with this idea, ectopically introduced RG4 constructs were folded in E. coli but also disrupted growth and translation. Future studies will focus on determining the molecular mechanism of the machinery that unfolds RG4s in eukaryotic cells.