Pervasive translation of circular RNAs driven by short IRES-like elements

Circular RNAs (circRNAs) are a class of abundant RNAs with ambiguous function. Although some circRNAs can be translated through IRES driven mechanisms, the scope and functions of circRNA translation are unclear because endogenous IRESs are rare. To determine the prevalence and mechanism of circRNA translation, we developed a cell-based system to screen random sequences and identified 97 overrepresented AU-rich hexamers (>2% of all hexamers) that drive cap-independent translation of circRNAs. These IRES-like short elements are significantly enriched in circRNAs and sufficient to drive circRNA translation. We further identified multiple trans-acting factors that bind these IRES-like short elements to initiate translation. Using mass-spectrometry data, hundreds of circRNA-coded peptides were identified, most of which have low abundance due to rapid degradation. As judged by mass-spectrometry, 50% of translatable endogenous circRNAs undergo rolling circle translation, several of which were experimentally validated by western blotting. Consistently, the mutation of the IRES-like short element in one circRNA reduced its translation. Collectively, our findings suggest a pervasive translation of circRNAs, providing profound implications in circRNA function. Overall design: A library of random 10-nt sequences was inserted before the start codon of circRNA-coded GFP, which was transfected into 293T cells to generate circRNAs that can be translated into intact GFP. The cells with active circRNA translation (i.e. green cells) can be recovered with fluorescence activated cell sorting (FACS), and the inserted decamers can be subsequently sequenced to identify the IRES-like elements that drive circRNA translation.