The power of cooperation: Experimental and computational approaches in the functional characterization of bacterial sRNAs Jens Georg, David Lalaouna, Shengwei Hou, Steffen Lott, Isabelle Caldelari, Stefano Marzi, Wolfgang Hess, Pascale Romby To cite this version: Jens Georg, David Lalaouna, Shengwei Hou, Steffen Lott, Isabelle Caldelari, et al.. The power ofco- operation: Experimental and computational approaches in the functional characterization of bacterial sRNAs. Molecular Microbiology, Wiley, 2019, 113, pp.603 - 612. 10.1111/mmi.14420. hal-02429851 HAL Id: hal-02429851 https://hal.archives-ouvertes.fr/hal-02429851 Submitted on 28 Oct 2020 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Received: 27 September 2019 | Revised: 30 October 2019 | Accepted: 6 November 2019 DOI: 10.1111/mmi.14420 MICROREVIEW The power of cooperation: Experimental and computational approaches in the functional characterization of bacterial sRNAs Jens Georg 1 | David Lalaouna 2 | Shengwei Hou 1 | Steffen C. Lott1 | Isabelle Caldelari 2 | Stefano Marzi 2 | Wolfgang R. Hess 1,3 | Pascale Romby 2 1Faculty of Biology, Genetics and Experimental Bioinformatics, University of Abstract Freiburg, Freiburg, Germany Trans-acting small regulatory RNAs (sRNAs) are key players in the regulation of gene 2 Architecture et Réactivité de l’ARN, CNRS, expression in bacteria. There are hundreds of different sRNAs in a typical bacte- Université de Strasbourg, Strasbourg, France 3Freiburg Institute for Advanced Studies, rium, which in contrast to eukaryotic microRNAs are more heterogeneous in length, University of Freiburg, Freiburg, Germany sequence composition, and secondary structure. The vast majority of sRNAs func- Correspondence tion post-transcriptionally by binding to other RNAs (mRNAs, sRNAs) through rather Wolfgang R. Hess, Faculty of Biology, short regions of imperfect sequence complementarity. Besides, every single sRNA Genetics and Experimental Bioinformatics, University of Freiburg, Schänzlestr. 1, may interact with dozens of different target RNAs and impact gene expression either D-79104 Freiburg, Germany. negatively or positively. These facts contributed to the view that the entirety of the Email: [email protected] regulatory targets of a given sRNA, its targetome, is challenging to identify. However, Pascale Romby, Architecture et Réactivité recent developments show that a more comprehensive sRNAs targetome can be de l’ARN, CNRS, Université de Strasbourg, UPR9002, Strasbourg, France. achieved through the combination of experimental and computational approaches. Email: [email protected] Here, we give a short introduction into these methods followed by a description of Present address two sRNAs, RyhB, and RsaA, to illustrate the particular strengths and weaknesses of Shengwei Hou, Department of Biological these approaches in more details. RyhB is an sRNA involved in iron homeostasis in Sciences, University of Southern California, Los Angeles, CA, USA Enterobacteriaceae, while RsaA is a modulator of virulence in Staphylococcus aureus. Using such a combined strategy, a better appreciation of the sRNA-dependent regu- Funding information Deutsche Forschungsgemeinschaft, Grant/ latory networks is now attainable. Award Number: GE 3159/1-1; Federal Ministry of Education and Research, Grant/ KEYWORDS Award Number: Partner grant 031L0106B CopraRNA, MAPS, post-transcriptional regulation, sRNAs, Staphylococcus aureus and grant RNAPronet 031L01; FRIAS-USIAS; Agence Nationale de la Recherche, Grant/ Award Number: ANR-16-CE11-007-01, ANR-10-LABX-0036, ANR-17-EURE-; EuropEuropean Union’s Horizon 2020 research and innovation programme, Grant/ Award Number: Marie Skłodowska-Curie Grant Agreement No. 753137 © 2019 The Authors. Molecular Microbiology published by John Wiley & Sons Ltd This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Jens Georg and David Lalaouna shared first authors. Molecular Microbiology. 2020;113:603–612. wileyonlinelibrary.com/journal/mmi | 603 604 | GEORG ET AL. 1 | THE CHALLENGE OF IDENTIFYING to enrich and identify the targetome of an sRNA (for a review, see THE REGULATORY TARGETS OF BACTERIAL (Saliba, C Santos, & Vogel, 2017)). Here, we will primarily focus on SRNAS MAPS (MS2-affinity purification coupled with RNA sequencing) and RIL-seq (RNA interaction by ligation and sequencing). Bacteria experience various metabolic and stress conditions they need In MAPS, the sRNA of interest is tagged usually at its 5′ end with to respond rapidly. Small trans-acting regulatory RNAs (sRNAs), which the MS2 RNA aptamer and expressed in bacteria. Cytoplasmic ex- frequently but not always are noncoding, have been found at the heart tracts are purified by affinity chromatography followed by the se- of regulatory pathways that allow bacteria to regulate virulence gene quencing of the enriched MS2–sRNA–RNA complexes (Lalaouna, expression, respond to stresses, sense the population density, modu- Prévost, Eyraud, & Massé, 2017; Lalaouna, Desgranges, Caldelari, late the cell surface composition, and adjust their metabolism (Carrier, & Marzi, 2018). The main advantages of this approach include the Lalaouna, & Massé, 2018; Desgranges, Marzi, Moreau, Romby, & sensitive detection of poorly expressed targets, the discrimination Caldelari, 2019; Holmqvist & Wagner, 2017; Radoshevich & Cossart, between direct and indirect targets, and finding sRNA targets that 2018). Therefore, it is of utmost interest to identify the RNA targets of form base pairings (Lalaouna, Carrier, et al., 2015). The approach was these sRNAs, as many bacteria express hundreds of different sRNAs applied successfully to many sRNAs from Gram-negative bacteria from differentially regulated genes. These sRNAs are very heteroge- such as E. coli and Salmonella Typhimurium, and from Gram-positive neous in length, sequence composition, and secondary structure. They bacteria like S. aureus (see below). can originate from their own genes or may be processed from the 5′ or 3′ To achieve a genome wide analysis of sRNA–RNA interactions, UTRs of protein-coding genes (Chao et al., 2017; Lalaouna et al., 2019). both CLASH (cross linking, ligation, and sequencing of hybrids) and Some sRNAs, such as ArcZ in Escherichia coli (Mandin & Gottesman, RIL-seq rely on the association of sRNAs–targets with RNA-binding 2010) or RprA in Salmonella enterica (Papenfort, Espinosa, Casadesus, & proteins (Melamed et al., 2016; Waters et al., 2017). In this application Vogel, 2015) are even further processed by RNase E, which resulted in of the CLASH method, sRNA–target hybrids bound by RNase E were different sRNA fragments. Although this processing is essential for the explored (Waters et al., 2017), while RIL-seq investigated interactions sRNA regulatory functions, it is not yet known whether the targets set of RNAs associated with the RNA chaperone Hfq (Melamed et al., of the sRNA is changed (Chao et al., 2017). In Bacillus subtilis, it has been 2016). In both methods, the tagged proteins and the bound RNAs are described that an additional RNase Y-dependent processing of RoxS purified by affinity chromatography, then the RNA hybrids are ligated expanded the repertoire of its target mRNAs (Durand et al., 2015). and the resulting RNA chimeras are sequenced. RIL-seq has expanded These sRNAs may not only interact with mRNAs but also with the ensemble of known targets of sRNAs in E. coli and showed the other sRNAs, tRNA precursors, or with proteins. However, the largest dynamics of the regulatory networks under various stress conditions class of bacterial sRNAs frequently targets different mRNAs that are (Melamed et al., 2016). These two interactome methods are not lim- often functionally related. The sRNA–mRNA interaction relies on base ited to a specific sRNA and can simultaneously identify a great number pairings (including Watson–Crick and G–U) between complementary of RNA–RNA interactions. The RIL-seq approach has been restricted sequences stretches in the two molecules. To complicate things fur- until now to E. coli as Hfq in several Gram-positive bacteria has lim- ther, these RNA sequence elements involve usually short (between ited RNA chaperone activity (Zheng, Panja, & Woodson, 2016). Only in 8 and less than 50 nts) and noncontinuous base pairings, and can be Listeria monocytogenes, Hfq has been found as a key partner of sRNAs involved in alternative intramolecular secondary structures competing (Nielsen et al., 2010). As Listeria expresses numerous sRNAs involved with the intermolecular interaction. Moreover, interacting proteins in virulence (Cerutti et al., 2017; Toledo-Arana et al., 2009), the RIL- such as the RNA chaperones Hfq (Dos Santos, Arraiano, & Andrade, Seq approach using Hfq might be an appropriate strategy to simul- 2019) and ProQ (Smirnov et al., 2016) frequently mediate these in- taneously probe RNA–sRNA interactions involved in stress