Discovery of new RNA classes and global RNA-binding proteins Alexandre Smirnov 1 , Cornelius Schneider 2 , Jens Ho ¨r 2 and Jo ¨ rg Vogel 2,3 The identification of new RNA functions and the functional annotation of transcripts in genomes represent exciting yet challenging endeavours of modern biology. Crucial insights into the biological roles of RNA molecules can be gained from the identification of the proteins with which they form specific complexes. Modern interactome techniques permit to profile RNA–protein interactions in a genome-wide manner and identify new RNA classes associated with globally acting RNA-binding proteins. Applied to a variety of organisms, these methods are already revolutionising our understanding of RNA-mediated biological processes. Here, we focus on one such approach — Gradient sequencing or Grad-seq — which has recently guided the discovery of protein ProQ and its associated small RNAs as a new domain of post-transcriptional control in bacteria. Addresses 1 UMR7156, CNRS, University of Strasbourg, 28, rue Goethe, FR-67083 Strasbourg, France 2 RNA Biology Group, Institute of Molecular Infection Biology, University of Wu ¨ rzburg, Josef-Schneider-Straße 2, D-97080 Wu ¨ rzburg, Germany 3 Helmholtz Institute for RNA-based Infection Research (HIRI), Josef- Schneider-Straße 2, D-97080 Wu ¨ rzburg, Germany Corresponding author: Current Opinion in Microbiology 2017, 39:152–160 This review comes from a themed issue on Bacterial systems biology Edited by Christoph Dehio and Dirk Bumann For a complete overview see the Issue and the Editorial Available online 24th November 2017 https://doi.org/10.1016/j.mib.2017.11.016 1369-5274/ã 2017 Published by Elsevier Ltd. Introduction Since the central dogma of molecular biology placed RNA as an informational intermediate between DNA and protein, it has not ceased to be corrected and nuanced to accommodate an ever-growing diversity of biological roles played by RNA in living cells. High-throughput sequencing and functional genomics methods have accel- erated the discovery of new noncoding RNA classes over the last decade, resulting in an intricate panorama of new RNA functions at nearly every stage of gene expres- sion [1,2]. The role of an RNA molecule in the cell is dictated by its sequence and structure. However, this information alone is often insufficient to predict RNA functions. This is especially true in bacteria which employ a variety of highly dissimilar yet mechanistically analogous noncod- ing RNAs to regulate their gene expression [3,4]. Adding to this complexity, small regulatory RNAs (sRNAs) can be embedded in other transcripts, giving rise to genomic loci with multiple functional outputs and overlapping annotations. For example, the 3 0 UTRs, and less often 5 0 UTRs or ORFs, of bacterial protein-coding genes may serve as a reservoir of sRNAs [5–7]. Some of these sRNAs possess their own promoters while others are generated by endonuclease-dependent processing of host mRNAs. Spotting and disentangling such cases by simple sequence and structure analysis, even helped by tran- scriptomic data, still meets with limited success. A different way to analyse RNA functions is to look at their associations with cognate RNA-binding proteins (RBPs) that are directly involved with their molecular activities [8]. This approach becomes particularly powerful when many RNAs interact with the same protein, which permits to enact the guilt-by-association logic to infer their function. For instance, in bacteria, sRNAs can differ in their origin, genomic organisation, biogenesis, vary in length by up to an order of magnitude and share practically no sequence similarity. However, common binding to the conserved RNA chaperone Hfq specifies many of them as trans-acting base-pairing riboregulators that typically modulate the sta- bility and/or translation of multiple target mRNAs [9,10]. To fully take advantage of this principle, it is important to develop robust genome-wide techniques permitting to identify groups of functionally related transcripts com- monly associated with RNA-binding hub proteins. In this review, we will first give a brief overview of general approaches exploiting information about RNA–protein interactions to address the functions of RNAs and RBPs. Subsequently, we will discuss the recently developed Grad-seq approach which offers the unique ability to directly detect groups of similarly behaving RNAs and guides the discovery of new globally acting RBPs in both model and poorly studied microbes. Experimental approaches for the global analysis of RNA–protein interactions There are three conceptually different groups of ap- proaches to detect RNA–protein interactions (Figure 1). Available online at www.sciencedirect.com ScienceDirect Current Opinion in Microbiology 2017, 39:152–160 www.sciencedirect.com