Small-Molecule-Binding THEA S. LOTZ1 and BEATRIX SUESS1 1Synthetic Genetic Circuits, Department of Biology, TU Darmstadt, 64287 Darmstadt, Germany

ABSTRACT RNA is a versatile capable of moieties is imperative in enabling RNA to function as a transferring information, taking on distinct three-dimensional biological catalyst, regulator, or structural scaffold. shapes, and reacting to ambient conditions. RNA molecules In the last 15 years, many different types of regula- utilize a wide range of mechanisms to control expression. tory RNA molecules have been discovered in nature. An An example of such regulation is riboswitches. Consisting exclusively of RNA, they are able to control important metabolic abundance of riboswitches, , RNA thermom- processes, thus providing an elegant and efficient RNA-only eters, and short and long noncoding have been regulation system. Existing across all domains of life, found in all three domains of life. However, plenty of riboswitches appear to represent one of the most highly questions regarding how RNA influences cell physio- conserved mechanisms for the regulation of a broad range of logy, differentiation, and development still remain un- biochemical pathways. Through binding of a wide range of answered and the subject of extensive research (1). There small-molecule ligands to their so-called , are several advantages when is regulated riboswitches undergo a conformational change in their downstream “expression platform.” In consequence, the pattern via RNA alone as opposed to regulation in combination of gene expression changes, which in turn results in increased or with . It allows (i) faster regulatory responses, decreased production. Riboswitches unite the sensing (ii) easier transfer of a single-step genetic control element and transduction of a signal that can directly be coupled to to other organisms, and (iii) flexible combination with the of the cell; thus they constitute a very potent different downstream readout platforms for a maximum regulatory mechanism for many organisms. Highly specific of regulatory outputs (2). in vivo RNA-binding domains not only occur but can also be Riboswitches are protein-independent, RNA-based evolved by means of the SELEX (systematic evolution of ligands — by exponential enrichment) method, which allows in vitro gene regulatory elements short, structured RNA ele- selection of against almost any ligand. Coupling of ments able to regulate gene expression in response to these aptamers with an expression platform has led to the binding a small-molecule ligand. Their use allows tem- development of synthetic riboswitches, a highly active research poral and dosage control over gene expression. Follow- field of great relevance and immense potential. The aim of this ing the discovery and validation of the first in review is to summarize developments in the riboswitch field over 2002 (3), they have been the subject of intense research. the last decade and address key questions of recent research.

Received: 11 February 2018, Accepted: 11 May 2018, INTRODUCTION Published: 3 August 2018 Editors: Gisela Storz, Division of Molecular and Cellular Biology, Traditionally, the functional role of RNA was thought Eunice Kennedy Shriver National Institute of Child Health and to be restricted to transferring genetic information Human Development, Bethesda, MD; Kai Papenfort, Department of Biology I, Microbiology, LMU Munich, Martinsried, Germany from DNA to protein. However, the discovery of RNA Citation: Lotz TS, Suess B. 2018. Small-molecule-binding elements mediating gene control, chemical reaction riboswitches. Microbiol Spectrum 6(4):RWR-0025-2018. catalysis, and signal transduction has changed this per- doi:10.1128/microbiolspec.RWR-0025-2018. ception fundamentally. Its ability to form complex three- Correspondence: Beatrix Suess, [email protected] © 2018 American Society for Microbiology. All rights reserved. dimensional structures that precisely present chemical

ASMscience.org/MicrobiolSpectrum 1 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Lotz and Suess

Their mechanism of action and importance for their typically a metabolite. They consist of two domains, host organism have been explored (4–6), along with the the so-called aptamer domain and the expression plat- generation of new, synthetic riboswitches by genetic form. The aptamer domain selectively recognizes its engineers (7, 8). small-molecule ligand while at the same time discrimi- In this review, we have compiled information on all nating against closely related variants of the ligand. the natural riboswitches discovered so far and discuss Binding of the ligand to the aptamer domain leads to their mechanisms, occurrence, and potential to develop structural changes in the following expression platform, genetic control elements for genetic analyses and syn- inducing altered expression of the downstream mRNA. thetic biology in the future. The changes in gene expression typically include regu- lating transcription termination or initiation (Fig. 1), more rarely, -mediated mRNA degra- RIBOSWITCHES—LOCATION, MECHANISM, dation or the control of splicing. AND DISTRIBUTION Riboswitches are widely distributed throughout the Riboswitches are highly structured RNA sequence ele- bacterial world and are increasingly found in ments typically located in the 5′ and . They are divided into so-called classes, (5′ UTR) of many bacterial mRNAs that control a groups of riboswitches that respond to the same ligand plethora of metabolic processes. They act as molecular and that show a similar conserved core structure. There switches regulating gene expression via conformational can be more than one class of riboswitches binding changes in their three-dimensional structure upon direct to the same ligand. However, the classes then have dis- interaction (“binding”) with a specific , tinctly different core structures. For example, among

FIGURE 1 Common mechanism of riboswitches in . (A) Regulation of translation initiation: In the absence of the ligand, a stem-loop structure is formed between the aptamer domain and a sequence complementary to the Shine-Dalgarno (SD) sequence. Thus, the SD sequence is accessible for 30S binding, and translation initiation occurs. As a consequence of ligand binding (pentagon) and the folding of the aptamer domain, an alternative stem-loop is formed, which sequesters the SD sequence, and the binding of the 30S ribosomal subunit is blocked. (B) Regulation of transcription termination: The aptamer domain is followed by a sequence complementary to the 3′ part of the aptamer and a U stretch. In the absence of the ligand, the complementary 3′ part is base-paired with the aptamer, forming a terminator structure. Thus, RNA polymerase (RNAP) dissociates and transcription is blocked. Upon ligand binding, terminator structure formation is inhibited and transcription can proceed, resulting in expression of the reporter gene.

2 ASMscience.org/MicrobiolSpectrum Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Small-Molecule-Binding Riboswitches

the prequeuosine-1 (preQ1) binding riboswitches, there system within an organism. This is concurrent with the are major differences between classes, as, e.g., preQ1-I now widely accepted RNA world theory, which pro- riboswitches are small (25 to 45 long) poses a phase of evolution where there were no proteins whereas preQ1-II have a larger, more complex core and DNA and all their current roles were carried out by consensus sequence (average, 58 nucleotides). Between RNA on its own. Modern riboswitches could be descen- the two classes, there are no detectable structure or dants of these ancient regulatory systems (6, 27, 28). sequence similarities (9, 10). In the meantime, close to 40 riboswitch classes have been identified and many more await discovery. The variety of ligands ranges THE TPP RIBOSWITCH—AN EXAMPLE from vitamins, amino acids, and nucleotides to second ACROSS ALL DOMAINS OF LIFE messengers and a number of ions. TPP-binding riboswitches belong to the most widely Several excellent reviews about riboswitches have distributed riboswitch class. They are found in all do- been published in recent years (4, 11–13). For a com- mains of life; the structure of their aptamer domain is prehensive overview, we have compiled a table (Table 1) highly conserved. Important differences between the TPP summarizing all riboswitch classes described to date. In riboswitches can be found in their regulation mecha- addition to the recognized ligands, we list the commonly nisms. In bacteria, ligand binding to the aptamer domain regulated and the main organisms in which the leads to changed transcription termination and transla- riboswitches occur. In addition, we include the year of tion initiation, respectively. Occasionally, an organism discovery of the ligand-riboswitch pair and the first de- also uses both mechanisms. Escherichia coli, for exam- scription of a three-dimensional structure if available. ple, controls the thiM via a TPP riboswitch reg- Albeit many RNA motifs have been justly suspected to ulated by translation initiation, whereas the thiC operon act as riboswitches since their discovery (14, 15), some is regulated both at the translational and transcriptional “orphan riboswitches” have not yet been assigned a level (17, 29). In eukarya, TPP riboswitches are associ- ligand. For this reason, the initial reference column of ated with splicing or alternative splicing events. Notably, riboswitches in Table 1 lists the first publication in which the TPP riboswitch from land is the only example a riboswitch could be successfully linked to its specific of gene control by means of a 3′ UTR-positioned ribo- ligand. switch in any organism to date (23). Despite the fact that riboswitches are widely distrib- Several atomic-resolution structures of the TPP- uted, some riboswitch classes are much more prevalent binding domain have been solved (30–32). The structure than others. The pyrophosphate (TPP) ribo- reveals a complex folded RNA with two subdomains. switch is the most widely distributed class and the only One subdomain forms an intercalation pocket for the one known to date to occur in all three domains of life: moiety of TPP (Fig. 2, J5-4), whereas another Bacteria (16, 17), Archaea (18), and Eukarya (fungi [19, subdomain forms a wider pocket that employs bivalent 20], plants [19, 21–24], and algae [25]). Cobalamin metal ions and water molecules to make bridging con- riboswitches (AdoCbl) belong to another widely dis- tacts to the pyrophosphate moiety of the ligand (J3-2). tributed riboswitch class and are located in the 5′ UTRs The two pockets are positioned to function as a molec- of -related genes across several strains of ular measuring device that recognizes TPP in an extended bacteria. In contrast, there are riboswitch classes that are conformation (Fig. 2). The central thiazole moiety is not extremely rare, corresponding to either recently evolved recognized by the RNA, which explains why the anti- pathways or such that are on the way to extinction microbial compound pyrithiamine pyrophosphate tar- (12). For instance, only four examples of the 2′-dG-I gets this riboswitch and downregulates the expression riboswitch class (sensing 2′-) have been of thiamine metabolic genes. The stabilization of the found, all of which occur in Mesoplasma florum (26). RNA structure by the natural ligand (but also its drug- Remarkably, a look across all known classes of ribo- like analog) is then transferred to the expression plat- switches reveals that most of them bind selectively to form, which then controls the synthesis of the proteins second-messenger signal molecules like cyclic di-GMP or coded by the mRNA. to coenzymes like TPP or S-adenosylmethionine (SAM). Although there are different genes regulated by TPP Both of these molecules can be derived from RNA nu- riboswitches, they are commonly associated with thia- cleotides or their precursors, implying a heavy reliance mine metabolism or transport (33). TPP is an active form of RNA on RNA-like substances or, in other words, of vitamin B1, an essential participant in many protein- an independence from proteins as part of a regulatory catalyzed reactions. TPP itself is indirectly involved in

ASMscience.org/MicrobiolSpectrum 3 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 4 TABLE 1 Naturally occurring riboswitches Suess and Lotz

Ligand Regulated gene(s) Distribution; organism Year first Reference(s) Riboswitch Ligand group or pathwaysa in which described firstb described Reference(s) (3D)c PDB ID AdoCbl Coenzyme Transport and Extremely widely distributed; 2002 28 86 4GMA (vitamin B12) of cobalamin or similar Escherichia coli metabolites; btuB and cob operon TPP Coenzyme Transport and biosynthesis Extremely widely distributed; 2002 16, 17, 25 30–32 2HOJ, 2GDI, (vitamin B1) of thiamine; thi operon E. coli, , 2CKY Neurospora crassa, Arabidopsis thaliana, and Volvox carteri FMN Flavin mononucleotide Coenzyme Biosynthesis and transport Extremely widely distributed; 2002 3 87, 88 3F2Q, 2YIF (vitamin B2) of riboflavin; rfn operon B. subtilis , Purine metabolism; xpt-pbuX Bacilli, clostridia, etc.; B. subtilis 2003 52 89 2EES derivative and ydhL L-Lysine Amino acid Lysin synthesis, catabolism, Widely distributed; B. subtilis 2003 22 90, 91 3D0U, 3DIL and transport; lysC SAM-I S-Adenosylmethionine Coenzyme Sulfur metabolism Extremely widely distributed; 2003 92 54, 93 2GIS, 3IQN (SAM) (biosynthesis of cysteine, B. subtilis , and SAM) AqCbl Aquacobalamin Coenzyme Cobalamin biosynthesis; Widely distributed; Salmonella 2004 94 86 4FRN cob operon enterica serovar Typhimurium GlmS Glucosamine-6-phosphate Others glmS Widely distributed; B. subtilis 2004 36 95, 96 2NZ4, 2Z75 Glycine Glycine Amino acid Glycine catabolism; Widely distributed; B. subtilis 2004 97 98 3OWW gcvT operon SAM-II SAM Coenzyme Methionine biosynthesis; Widely distributed; 2005 99 100 2QWY metA and metC Agrobacterium tumefaciens SAM-III SAM Coenzyme SAM synthetase; metK Enterococcus, Streptococcus, 2006 101 102 3E5C and Lactococcus species Mg2+ -II Magnesium ions, Ion mgtA and mgtL Gammaproteobacteria; 2006 103, 104 proline S. enterica

ASMscience.org/MicrobiolSpectrum Mg2+ -I Magnesium ions Ion Magnesium homeostasis Widely distributed; B. subtilis 2007 105 106 3PDR 2′-dG-I 2′-Deoxyguanosine Nucleotide Subunit β of Mesoplasma florum 2007 26 107 3SKI (mfl motif) derivative reductase

PreQ1-I type Prequeuosine-1 Nucleotide biosynthesis and Widely distributed; B. subtilis, 2007 9, 108–110 111 3Q50 1–3 derivative transport; queCDEF operon E. coli, and Shigella dysenteriae and queT

PreQ1-II Prequeuosine-1 Nucleotide Queuosine biosynthesis; Bacilli and clostridia; 2007 9, 10 derivative COG4708 (proposed Streptococcus pneumoniae transporter) Moco Molybdenum cofactor Coenzyme Molybdate transporters, Widely distributed; E. coli 2008 112 Moco biosynthesis, and Moco-containing proteins SAH S-Adenosylhomocysteine Coenzyme SAM recycling (converting Widely distributed; 2008 113 114 3NPN, (SAH) SAH to L-methionine); metH Pseudomonas syringae 3NPQ SAM-IV SAM Coenzyme Sulfur metabolism Actinomycetes; Mycobacterium 2008 115 tuberculosis Wco Tungsten cofactor Coenzyme Riboswitch suspected Widely distributed 2008 112 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 ASMscience.org/MicrobiolSpectrum c-di-GMP-I 3′-5′-Cyclic-di-GMP Signaling Various cellular processes Widely distributed; Clostridium 2008 116 53, 117 3IRN, 3IWN molecule difficile and Vibrio cholerae SAM-V SAM Coenzyme Sulfur metabolism Marine alphaproteobacteria; 2009 118 Pelagibacter ubique c-di-GMP-II 3′-5′-Cyclic-di-GMP Signaling c-di-GMP biosynthesis, C. difficile 2010 119 120 3Q3Z molecule degradation, and signaling SAM-I/IV SAM Coenzyme Marine phyla 2010 15 SAM-SAH SAM, SAH Coenzyme SAM synthetase; metK Alphaproteobacteria; 2010 15 Roseobacter sp. THF Tetrahydrofolate Coenzyme Folate transport and Firmicutes, clostridia, 2010 121 122 3SD3, 4LVV biosynthesis; folT, folE, and lactobacilli folC, and folQPBK L-Glutamine Amino acid metabolism Rare cyanobacteria and marine 2011 123 124 5DDP metagenomic sequences; Synechococcus elongatus Fluoride Fluoride ions Ion Fluoride-sensitive enzymes Extremely widely distributed; 2012 125 126 4ENC (crcB motif) and fluoride exporter; CrcB Pseudomonas syringae c-di-AMP 3′-5′-Cyclic-di-AMP Signaling Cell wall metabolism, osmotic B. subtilis 2013 14, 127 128 4QLM (ydaO motif) molecule stress, and sporulation

PreQ1-III Prequeuosine-1 Nucleotide Transport of queuosine Rare; Faecalibacterium 2014 129 130 4RZD derivative derivatives; queT prausnitzii and Shigella dysenteriae Mn2+ Manganese ions Ion Metal ion household, Widely distributed; 2015 131 132 4Y1I (yybP motif) tellurium resistance, Lactococcus lactis, E. coli, and cation-transport and B. subtilis ATPase; yybP-ykoY cAMP-GMP 3′-5′-Cyclic AMP-GMP Signaling Extracellular electron Rare; Geobacter 2015 133, 134 135 4YAZ molecule transfer; pgcA sulfurreducens NiCo or ions Ion Heavy metal ion sensing Rare; Clostridium scindens 2015 136 136 4RUM and Clostridium botulinum ZTP (pfl motif) 5-Aminoimidazole-4- Signaling De novo purine biosynthesis Widely distributed; 2015 137 138 4ZNP carboxamide riboside molecule and one-carbon metabolism Streptococcus thermophilus 5′-triphosphate Aza-aromatic Aza-aromatics Others yjdF Firmicutes; B. subtilis 2016 139 ml-oeueBnigRiboswitches Small-Molecule-Binding (yjdF motif) Guanidine-I Guanidine Others Urea carboxylases and Widely distributed; 2016 140 141 5T83 (ykkC motif) multidrug efflux pumps B. subtilis and E. coli (e.g., EmrE and SugE) 2′-dG-II 2′-Deoxyguanosine Signaling Phosphate transporter, Metagenomic sequence data 2017 47 molecule phospholipase D, endonuclease I, and signal receiver domain Guanidine-II Guanidine Others SMR efflux pumps; Rather rare, mostly 2017 142 143 5NDI (mini-ykkC) EmrE and SugE Guanidine-III Guanidine Others SMR efflux pumps; Rare, mostly actinobacteria 2017 144 145 5NWQ (ykkC-III) EmrE and SugE

aMost riboswitches regulate a multitude of genes; here we try to list the ones described first. bDistribution is based on reference 12; in addition, the table lists the organisms in which the riboswitches were first described. cPublication of the first three-dimensional structure of the riboswitch. 5

Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Lotz and Suess

FIGURE 2 Structure of the TPP riboswitch. (A) Schematic depiction of the secondary structure of the thiM riboswitch with and without TPP (marked in pink). TPP stabilizes the P1-P1′ helix, which leads to secondary structure changes. The formation of the expression platform follows as a consequence, so that the Shine-Dalgarno (SD) sequence is se- questered in another stem, inhibiting any further gene expression. (B) The X-ray crystal structure of the E. coli thiM aptamer bound to TPP (black sticks, center) and Mg2+ ions (black spheres). PDB ID 2GDI (31); annotations on the structures refer to helices (P) and junctions (J). Adapted from reference 85 with permission. several metabolic steps of the electron transport chain, ingested (38). As these bacteria live in two very different contributes to energy extraction from carbohydrate environments, i.e., the human body and marine con- sources (34, 35), and acts as a coenzyme in the conver- ditions, they need to regulate their metabolism according sion of glucose to ATP. Therefore, it is crucial for cells to environmental conditions. Its adenine riboswitch acts that the metabolism is correctly controlled and regu- in a translational manner and can sense temperature and lated. The fact that riboswitches carry out such a vital ligand concentration to mediate efficient gene expres- function throughout the kingdoms of life stresses their sion regulation over the range of physiologically rele- versatility and reliability across the generations of evo- vant temperatures, allowing the organism to adapt to the lution already undergone. differing temperatures experienced in the free-living state and within the human intestine (39). Other dual-input systems in nature favor a two-pronged approach to gene VERSATILITY OF RIBOSWITCHES regulation, such as the prokaryote Bacillus clausii. Two In nature, there are several combinatorial approaches different independent riboswitches that respond to SAM to solve complex regulatory problems. An excellent ex- and vitamin B12, respectively, can be found in the 5′ UTR ample is the glmS riboswitch discovered in Bacillus sub- of the metE mRNA. Analysis of this system showed that tilis, as it can respond to more than one ligand. It acts as it fits the theory behind a Boolean NOR gate, as tran- a ribozyme that self-cleaves once it binds to glucosamine- scription termination can be achieved in the presence of 6-phosphate. The glucosamine-6-phosphate-binding ribo- only one or both ligands (40). switch thus represents a classic negative feedback loop, Although riboswitches act as cis-regulatory elements as it regulates the gene responsible for its own produc- of their downstream mRNAs for the most part, some tion (glmS)(36). However, it has been shown that the combine gene expression in cis and trans. They are able glms riboswitch can also bind to glucose-6-phosphate; to act as processed, terminated riboswitches that operate yet this binding leads to an inhibition of ribozyme func- like a small RNA (41), as full-length transcripts contain- tion and stops subsequent self-cleavage. The riboswitch ing a binding site for sequestering an activated RNA- is thus able to react to the overall metabolic state of binding protein (42, 43) or a regulating part of antisense the cell and to regulate gene expression accordingly RNA, designating the antisense RNAs function (44). (37). Another example of a single riboswitch processing These trans-acting abilities of riboswitches further ex- two input signals is the adenine riboswitch encoded by pand their regulatory abilities. The ability of a ribo- the add gene from Vibrio vulnificus, a bacterium that switch to not only serve as a regulator of its associated causes severe gastrointestinal infections in humans when downstream gene but also to control gene expression

6 ASMscience.org/MicrobiolSpectrum Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Small-Molecule-Binding Riboswitches in trans makes the functional RNA-protein compari- RNA with single-stranded regions, which renders these son even more appropriate. It will be interesting to see more flexible and unstable so that they will degrade over whether future research uncovers more of these dual- time. The method involves incubation of structural or function riboswitches, which represent a complex, effi- functional RNAs over a long period of time (up to sev- ciently organized RNA regulation system. eral days) and the subsequent visualization of the bands by gel electrophoresis. In-line probing is also used to de- termine structural changes due to ligand binding. Ap- HOW TO FIND NEW RIBOSWITCHES plication of different ligand concentrations allows rapid All along, identification of new riboswitches has been estimation of the dissociation constant (Kd), making it a supported by and is nowadays almost exclusively carried method supplying information on riboswitch secondary out by tools. Software that used sequence structure (single- or double-stranded regions), nucleo- analysis for secondary structure and folding energy cal- tides involved in ligand binding, and binding affinity. culations and subsequently compared these values to Structural probing has the additional benefit that it can B. subtilis RNA motif variants was published as early as be carried out in aqueous solutions, which renders the 2004 (45). Since then, complex bioinformatics search resulting structures more likely to resemble their natural pipelines have been developed, which not only can pre- state of the RNA inside their organisms. Due to its low- dict novel riboswitch candidates (46) but can search tech approach and fast appraisal of results, many ribo- entire genome databases for the presence of known ri- switch publications first apply in-line probing to obtain boswitches or variants whose ligand-binding specificities information on the secondary structure of putative might have been altered (47). However, it is important riboswitches (52). to note that bioinformatics predictions cannot replace To get an overview of the general riboswitch struc- experimental validation of those switches. An early ex- ture, more complex methods than biochemical probing ample of an RNA structure correctly identified as a of the secondary structure must be employed. Small- riboswitch that did not fit in its predicted class is the angle X-ray scattering (SAXS) analysis (53, 54) as well adenine riboswitch. Attributed to the class of guanine as X-ray crystallography (55–57) and nuclear magnetic riboswitches (19), it contains a single nucleotide sub- resonance (NMR) spectroscopy (58, 59) have been ap- stitution in the conserved aptamer core that leads to plied successfully, but these approaches are laborious a change in ligand recognition. The riboswitch recog- and depend on specialized knowledge and experimen- nizes adenine as its ligand and discriminates against tal equipment. Each method is associated with specific guanine and other purine analogs with a high selectivity challenges. While SAXS analysis can deliver information of several orders of magnitude (48). This example shows on different conformations of a riboswitch in solution, that riboswitches, which might have evolved to have a it does not yield high-resolution insights. In contrast, the different ligand specificity but still retain much of their resolution of X-ray crystallographic analysis is very high, structural heritage, might be incorrectly assigned in bio- but as it only shows the details of one “frozen” confor- informatics searches. mation, this method can deliver an incomplete picture of all the different conformations possible in solution, particularly when analyzing ligand-free riboswitches. GETTING A GLIMPSE OF However, it must be emphasized that in recent years, RIBOSWITCH STRUCTURE the structure of the aptamer domains of many ribo- As structure is closely tied to function, and the bind- switches has been successfully determined by X-ray crys- ing of a specific ligand to the riboswitch induces a con- tallography, providing an interesting insight into the formational change, the study of riboswitch structures ligand-binding state of these regulators (see Table 1, last is pivotal. A challenge in structure determination of column). NMR spectroscopy has the scope to gather riboswitches is their intrinsic flexibility. In most cases, atomic-resolution information on different conforma- riboswitches without their ligand present can adopt sev- tional states in solution, but there are limits to the mo- eral different conformations, and only the binding of lecular size suitable for analysis, with riboswitches the ligand leads to the adoption of a rigid and rela- already close to or above the upper size threshold. tively inflexible structure (49). There are a number of As many structures to date only provide a “time- methods to determine the secondary structure of nucleic frozen” conformation of aptamer and ligand together, acids in general, e.g., in-line probing (50, 51). This type the process of ligand-induced folding of the riboswitch of probing assay makes use of the natural instability of itself generally remains unexplored. To tackle this chal-

ASMscience.org/MicrobiolSpectrum 7 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Lotz and Suess lenge, the method of single-molecule fluorescence reso- Efficient conditional gene expression can be obtained nance energy transfer has recently become a tool to using in vitro-selected aptamers binding to tetracycline, study the dynamics of the ligand-free riboswitch struc- neomycin, or ciprofloxacin (71–73). However, aptamers ture in particular (49, 60, 61). Its benefits include the can also be applied to control pre-mRNA splicing (74– analysis of one single riboswitch molecule only, which 76), microRNA processing (77, 78), or internal ribosome is advantageous when studying dynamic biomolecular entry site-mediated translation initiation (79). Several systems that can exist in multiple distinct conformations. independent approaches have sought to regulate gene In summary, riboswitch structure determination remains expression by ribozymes by coupling aptamer domains a complex issue. Although a plethora of methods are (theophylline,tetracycline,andneomycinaptamer)tovar- available, a full understanding of individual riboswitch ious ribozymes (hammerhead, HDV, and twister) (80, structure and its dynamics still requires application of a 81). These methods allow efficient induction and inhibi- combination of several of these methods. tion, respectively, of gene expression in bacteria (82), yeast (83), and even human cell lines (84), to name only a few. Currently, gaps in the riboswitch toolbox are being SYNTHETIC RIBOSWITCHES—NOVEL filled with new aptamer domains, new expression mod- MEANS FOR THE CONDITIONAL ules, and combinations thereof. In addition, advanced CONTROL OF GENE EXPRESSION in vitro and in vivo screening systems and improved An outstanding characteristic of riboswitches is that in silico and structure-based design approaches have RNA acts as both sensor and effector, demonstrating added to our riboswitch engineering principles and that a protein cofactor for sensing the ligand is not an methods, so that the riboswitches we generate should obligate requirement for regulation. Motivated by this, more closely resemble their natural counterparts. Con- researchers have engineered a versatile set of synthetic solidation of the knowledge gained in all of these and riboswitches by combining in vitro generated aptamer future experiments will create positive feedback and domains (using the SELEX method) with expression ultimately permit the design of even better tools that platforms (62, 63). These now represent promising and should finally allow standardized “off-the-shelf” com- powerful tools in synthetic biology. Several approaches ponents to be utilized in “plug-and-play” approaches for based on the assembly of individual riboswitch building the generation of synthetic genetic circuits of adjustable blocks de novo-designed into functional regulators have complexity, sophisticated biosensors, “intelligent” re- been realized. The best working synthetic riboswitches in sponsive metabolic pathways, and optimized diagnostic bacteria exploit the theophylline aptamer as their sensing and therapeutic tools. domain. Through the use of screening systems, synthetic riboswitches that can sequester the ribosome binding site depending on the presence of the ligand have been CONCLUDING REMARKS developed. A series of different variants allow the effi- Although there has been considerable research explor- cient application in a variety of bacteria (64, 65). Using ing new natural or synthetic riboswitches, it is still a an in silico approach combined with additional optimi- new field of study. Benefiting from the steady decrease zation steps, the same aptamer and the tetracycline ap- in cost and complexity of high-throughput analysis of tamer were used to develop synthetic riboswitches that entire genomes, rapid advancements in bioinformatics control transcription termination (66). Another elegant analysis, and growing understanding of the relation of strategy using a modular “mix-and-match” approach RNA structure and function, future discovery of new was applied to construct transcriptional switches. In this riboswitches will likely be less expensive and less time- work, expression platforms from natural riboswitches consuming. Despite the fact that plenty of highly abundant were combined with aptamer domains both derived from riboswitch classes have already been identified, Breaker riboswitches and selected in vitro (67, 68). Other ap- and colleagues predict that there could potentially be proaches tried to change the ligand specificity of natural thousands of additional undiscovered riboswitch classes, riboswitches (69, 70). albeit many of these are rare (12). In this review, we have In contrast to the rather rare examples of eukaryotic tried to provide a comprehensive overview of the timeline riboswitches in nature, various synthetic riboswitches of natural riboswitch discovery, as well as a summary have been developed for application in eukaryotes. The of the validated classes of riboswitches and their distin- easiest strategy is the simple insertion of small-molecule- guishing features to date, and conclude with an overview binding aptamers into the 5′ UTR of a gene of interest. of the progress in the field of synthetic riboswitches.

8 ASMscience.org/MicrobiolSpectrum Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Small-Molecule-Binding Riboswitches

ACKNOWLEDGMENTS 19. Sudarsan N, Barrick JE, Breaker RR. 2003. Metabolite-binding We thank Dr. Florian Groher for the preparation of Figure 1. This RNA domains are present in the genes of eukaryotes. RNA 9:644–647. work was supported by the Deutsche Forschungsgemeinschaft http://dx.doi.org/10.1261/rna.5090103. (SFB902/A2) and LOEWE CompuGene. 20. Cheah MT, Wachter A, Sudarsan N, Breaker RR. 2007. Control of alternative RNA splicing and gene expression by eukaryotic riboswitches. – REFERENCES Nature 447:497 500. http://dx.doi.org/10.1038/nature05769. 1. Morris KV, Mattick JS. 2014. The rise of regulatory RNA. Nat Rev 21. Kubodera T, Watanabe M, Yoshiuchi K, Yamashita N, Nishimura A, Genet 15:423–437. http://dx.doi.org/10.1038/nrg3722. Nakai S, Gomi K, Hanamoto H. 2003. Thiamine-regulated gene expres- sion of Aspergillus oryzae thiA requires splicing of the intron containing a 2. Etzel M, Mörl M. 2017. Synthetic riboswitches: from plug and pray riboswitch-like domain in the 5′-UTR. FEBS Lett 555:516–520. http://dx toward plug and play. Biochemistry 56:1181–1198. http://dx.doi.org .doi.org/10.1016/S0014-5793(03)01335-8. /10.1021/acs.biochem.6b01218. 22. Sudarsan N, Wickiser JK, Nakamura S, Ebert MS, Breaker RR. 3. Winkler WC, Cohen-Chalamish S, Breaker RR. 2002. An mRNA 2003. An mRNA structure in bacteria that controls gene expression by structure that controls gene expression by binding FMN. Proc Natl Acad binding lysine. Genes Dev 17:2688–2697. http://dx.doi.org/10.1101/gad Sci U S A 99:15908–15913. http://dx.doi.org/10.1073/pnas.212628899. .1140003. 4. Serganov A, Nudler E. 2013. A decade of riboswitches. Cell 152:17–24. 23. Wachter A, Tunc-Ozdemir M, Grove BC, Green PJ, Shintani DK, http://dx.doi.org/10.1016/j.cell.2012.12.024. Breaker RR. 2007. Riboswitch control of gene expression in plants by 5. Wachter A. 2014. Gene regulation by structured mRNA elements. splicing and alternative 3′ end processing of mRNAs. Cell 19:3437– – Trends Genet 30:172 181. http://dx.doi.org/10.1016/j.tig.2014.03.001. 3450. http://dx.doi.org/10.1105/tpc.107.053645. 6. Breaker RR. 2012. Riboswitches and the RNA world. Cold Spring Harb 24. Bocobza SE, Aharoni A. 2014. Small molecules that interact with RNA: Perspect Biol 4:a003566. http://dx.doi.org/10.1101/cshperspect.a003566. riboswitch-based gene control and its involvement in metabolic regulation 7. Wittmann A, Suess B. 2012. Engineered riboswitches: expanding in plants and algae. Plant J 79:693–703. http://dx.doi.org/10.1111/tpj researchers’ toolbox with synthetic RNA regulators. FEBS Lett 586: .12540. – 2076 2083. http://dx.doi.org/10.1016/j.febslet.2012.02.038. 25. Croft MT, Moulin M, Webb ME, Smith AG. 2007. Thiamine bio- 8. Wieland M, Ausländer D, Fussenegger M. 2012. Engineering of synthesis in algae is regulated by riboswitches. Proc Natl Acad Sci U S A ribozyme-based riboswitches for mammalian cells. Methods 56:351–357. 104:20770–20775. http://dx.doi.org/10.1073/pnas.0705786105. http://dx.doi.org/10.1016/j.ymeth.2012.01.005. 26. Kim JN, Roth A, Breaker RR. 2007. Guanine riboswitch variants 9. Roth A, Winkler WC, Regulski EE, Lee BW, Lim J, Jona I, Barrick from Mesoplasma florum selectively recognize 2′-deoxyguanosine. Proc JE, Ritwik A, Kim JN, Welz R, Iwata-Reuyl D, Breaker RR. 2007. A Natl Acad Sci U S A 104:16092–16097. http://dx.doi.org/10.1073/pnas riboswitch selective for the queuosine precursor preQ1 contains an un- .0705884104. – usually small aptamer domain. Nat Struct Mol Biol 14:308 317. http://dx 27. Gilbert W. 1986. Origin of life: the RNA world. Nature 319:618. .doi.org/10.1038/nsmb1224. http://dx.doi.org/10.1038/319618a0. 10. Meyer MM, Roth A, Chervin SM, Garcia GA, Breaker RR. 2008. 28. Nahvi A, Sudarsan N, Ebert MS, Zou X, Brown KL, Breaker RR. fi Con rmation of a second natural preQ1 aptamer class in Streptococcaceae 2002. Genetic control by a metabolite binding mRNA. Chem Biol bacteria. RNA 14:685–695. http://dx.doi.org/10.1261/rna.937308. 9:1043–1049. http://dx.doi.org/10.1016/S1074-5521(02)00224-7. 11. Batey RT. 2015. Riboswitches: still a lot of undiscovered country. 29. Ontiveros-Palacios N, Smith AM, Grundy FJ, Soberon M, Henkin RNA 21:560–563. http://dx.doi.org/10.1261/rna.050765.115. TM, Miranda-Ríos J. 2008. Molecular basis of gene regulation by the 12. McCown PJ, Corbino KA, Stav S, Sherlock ME, Breaker RR. 2017. THI-box riboswitch. Mol Microbiol 67:793–803. http://dx.doi.org/10 Riboswitch diversity and distribution. RNA 23:995–1011. http://dx.doi .1111/j.1365-2958.2007.06088.x. .org/10.1261/rna.061234.117. 30. Edwards TE, Ferré-D’Amaré AR. 2006. Crystal structures of the 13. Jones CP, Ferré-D’Amaré AR. 2017. Long-range interactions in thi-box riboswitch bound to thiamine pyrophosphate analogs reveal riboswitch control of gene expression. Annu Rev Biophys 46:455–481. adaptive RNA-small molecule recognition. Structure 14:1459–1468. http://dx.doi.org/10.1146/annurev-biophys-070816-034042. http://dx.doi.org/10.1016/j.str.2006.07.008. 14. Barrick JE, Corbino KA, Winkler WC, Nahvi A, Mandal M, Collins J, 31. Serganov A, Polonskaia A, Phan AT, Breaker RR, Patel DJ. Lee M, Roth A, Sudarsan N, Jona I, Wickiser JK, Breaker RR. 2004. 2006. Structural basis for gene regulation by a thiamine pyrophosphate- New RNA motifs suggest an expanded scope for riboswitches in bacterial sensing riboswitch. Nature 441:1167–1171. http://dx.doi.org/10.1038 genetic control. Proc Natl Acad Sci U S A 101:6421–6426. http://dx.doi /nature04740. .org/10.1073/pnas.0308014101. 32. Thore S, Leibundgut M, Ban N. 2006. Structure of the eukaryotic 15. Weinberg Z, Wang JX, Bogue J, Yang J, Corbino K, Moy RH, Breaker thiamine pyrophosphate riboswitch with its regulatory ligand. Science – RR. 2010. Comparative genomics reveals 104 candidate structured RNAs 312:1208 1211. http://dx.doi.org/10.1126/science.1128451. from bacteria, archaea, and their metagenomes. Genome Biol 11:R31. 33. Manzetti S, Zhang J, van der Spoel D. 2014. Thiamin function, me- http://dx.doi.org/10.1186/gb-2010-11-3-r31. tabolism, uptake, and transport. Biochemistry 53:821–835. http://dx.doi 16. Mironov AS, Gusarov I, Rafikov R, Lopez LE, Shatalin K, Kreneva .org/10.1021/bi401618y. RA, Perumov DA, Nudler E. 2002. Sensing small molecules by nascent 34. Franken JF, Stapert FP. 1954. Restoration of pyruvate breakdown RNA: a mechanism to control transcription in bacteria. Cell 111:747– in pigeon muscle homogenates impaired by thiamine deficiency. Bio- 756. http://dx.doi.org/10.1016/S0092-8674(02)01134-0. chim Biophys Acta 14:293–294. http://dx.doi.org/10.1016/0006-3002(54) 17. Winkler W, Nahvi A, Breaker RR. 2002. Thiamine derivatives bind 90178-X. messenger RNAs directly to regulate bacterial gene expression. Nature 35. Asakawa T, Wada H, Yamano T. 1968. Enzymatic conversion of 419:952–956. http://dx.doi.org/10.1038/nature01145. phenylpyruvate to phenylacetate. Biochim Biophys Acta 170:375–391. 18. Rodionov DA, Vitreschak AG, Mironov AA, Gelfand MS. 2002. http://dx.doi.org/10.1016/0304-4165(68)90017-2. Comparative genomics of thiamin biosynthesis in procaryotes. New genes 36. Winkler WC, Nahvi A, Roth A, Collins JA, Breaker RR. 2004. and regulatory mechanisms. J Biol Chem 277:48949–48959. http://dx Control of gene expression by a natural metabolite-responsive ribozyme. .doi.org/10.1074/jbc.M208965200. Nature 428:281–286. http://dx.doi.org/10.1038/nature02362.

ASMscience.org/MicrobiolSpectrum 9 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Lotz and Suess

37. Watson PY, Fedor MJ. 2011. The glmS riboswitch integrates signals 55. Scott WG, Finch JT, Grenfell R, Fogg J, Smith T, Gait MJ, Klug A. from activating and inhibitory metabolites in vivo. Nat Struct Mol Biol 1995. Rapid crystallization of chemically synthesized hammerhead RNAs 18:359–363. http://dx.doi.org/10.1038/nsmb.1989. using a double screening procedure. J Mol Biol 250:327–332. http://dx 38. Farmer JJ III. 1979. Vibrio (“Beneckea”) vulnificus, the bacterium .doi.org/10.1006/jmbi.1995.0380. associated with sepsis, septicaemia, and the sea. Lancet 2:903. http://dx 56. Gilbert SD, Montange RK, Stoddard CD, Batey RT. 2006. Structural .doi.org/10.1016/S0140-6736(79)92715-6. studies of the purine and SAM binding riboswitches. Cold Spring Harb 39. Loh E, Righetti F, Eichner H, Twittenhoff C, Narberhaus F. 2018. Symp Quant Biol 71:259–268. http://dx.doi.org/10.1101/sqb.2006.71 RNA thermometers in bacterial pathogens. Microbiol Spectr 6:RWR- .015. 0012-2017. http://dx.doi.org/10.1128/microbiolspec.RWR-0012-2017. 57. Edwards AL, Garst AD, Batey RT. 2009. Determining structures of 40. Sudarsan N, Hammond MC, Block KF, Welz R, Barrick JE, Roth RNA aptamers and riboswitches by X-ray crystallography. Methods Mol A, Breaker RR. 2006. Tandem riboswitch architectures exhibit complex Biol 535:135–163. http://dx.doi.org/10.1007/978-1-59745-557-2_9. gene control functions. Science 314:300–304. http://dx.doi.org/10.1126 58. Noeske J, Buck J, Fürtig B, Nasiri HR, Schwalbe H, Wöhnert J. 2007. /science.1130716. Interplay of ‘induced fit’ and preorganization in the ligand induced folding 41. Loh E, Dussurget O, Gripenland J, Vaitkevicius K, Tiensuu T, Mandin of the aptamer domain of the guanine binding riboswitch. Nucleic Acids P, Repoila F, Buchrieser C, Cossart P, Johansson J. 2009. A trans-acting Res 35:572–583. http://dx.doi.org/10.1093/nar/gkl1094. riboswitch controls expression of the virulence regulator PrfA in Listeria 59. Ottink OM, Rampersad SM, Tessari M, Zaman GJ, Heus HA, Wijmenga monocytogenes. Cell 139:770–779. http://dx.doi.org/10.1016/j.cell.2009 SS. 2007. Ligand-induced folding of the guanine-sensing riboswitch is con- .08.046. trolled by a combined predetermined induced fit mechanism. RNA 13:2202– 42. DebRoy S, Gebbie M, Ramesh A, Goodson JR, Cruz MR, van Hoof A, 2212. http://dx.doi.org/10.1261/rna.635307. Winkler WC, Garsin DA. 2014. Riboswitches. A riboswitch-containing 60. Haller A, Soulière MF, Micura R. 2011. The dynamic nature of RNA sRNA controls gene expression by sequestration of a response regulator. as key to understanding riboswitch mechanisms. Acc Chem Res 44:1339– Science 345:937–940. http://dx.doi.org/10.1126/science.1255091. 1348. http://dx.doi.org/10.1021/ar200035g. 43. Mellin JR, Koutero M, Dar D, Nahori MA, Sorek R, Cossart P. 2014. 61. Suddala KC, Walter NG. 2014. Riboswitch structure and dynamics by Riboswitches. Sequestration of a two-component response regulator by smFRET microscopy. Methods Enzymol 549:343–373. http://dx.doi.org a riboswitch-regulated noncoding RNA. Science 345:940–943. http://dx /10.1016/B978-0-12-801122-5.00015-5. .doi.org/10.1126/science.1255083. 62. Ellington AD, Szostak JW. 1990. In vitro selection of RNA molecules 44. Mellin JR, Tiensuu T, Bécavin C, Gouin E, Johansson J, Cossart P. that bind specific ligands. Nature 346:818–822. 2013. A riboswitch-regulated antisense RNA in Listeria monocytogenes. 63. Tuerk C, Gold L. 1990. Systematic evolution of ligands by exponential Proc Natl Acad Sci U S A 110:13132–13137. http://dx.doi.org/10.1073 enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science /pnas.1304795110. 249:505–510. 45. Bengert P, Dandekar T. 2004. Riboswitch finder—a tool for identi- 64. Topp S, Reynoso CM, Seeliger JC, Goldlust IS, Desai SK, Murat D, fication of riboswitch RNAs. Nucleic Acids Res 32(Web Server issue): Shen A, Puri AW, Komeili A, Bertozzi CR, Scott JR, Gallivan JP. 2010. W154–W159. http://dx.doi.org/10.1093/nar/gkh352. Synthetic riboswitches that induce gene expression in diverse bacterial 46. Chang TH, Huang HD, Wu LC, Yeh CT, Liu BJ, Horng JT. 2009. species. Appl Environ Microbiol 76:7881–7884. http://dx.doi.org/10.1128 Computational identification of riboswitches based on RNA conserved /AEM.01537-10. functional sequences and conformations. RNA 15:1426–1430. http://dx 65. Berens C, Suess B. 2015. Riboswitch engineering—making the all- .doi.org/10.1261/rna.1623809. important second and third steps. Curr Opin Biotechnol 31:10–15. http:// 47. Weinberg Z, Nelson JW, Lünse CE, Sherlock ME, Breaker RR. 2017. dx.doi.org/10.1016/j.copbio.2014.07.014. Bioinformatic analysis of riboswitch structures uncovers variant classes 66. Wachsmuth M, Findeiß S, Weissheimer N, Stadler PF, Mörl M. 2013. with altered ligand specificity. Proc Natl Acad Sci U S A 114:E2077– De novo design of a synthetic riboswitch that regulates transcription E2085. http://dx.doi.org/10.1073/pnas.1619581114. termination. Nucleic Acids Res 41:2541–2551 http://dx.doi.org/10.1093 48. Mandal M, Breaker RR. 2004. Adenine riboswitches and gene acti- /nar/gks1330. vation by disruption of a transcription terminator. Nat Struct Mol Biol 67. Ceres P, Garst AD, Marcano-Velázquez JG, Batey RT. 2013. Mod- 11:29–35. http://dx.doi.org/10.1038/nsmb710. ularity of select riboswitch expression platforms enables facile engineering 49. Liberman JA, Wedekind JE. 2012. Riboswitch structure in the ligand- of novel genetic regulatory devices. ACS Synth Biol 2:463–472. http://dx free state. Wiley Interdiscip Rev RNA 3:369–384. http://dx.doi.org/10.1002 .doi.org/10.1021/sb4000096. /wrna.114. 68. Ceres P, Trausch JJ, Batey RT. 2013. Engineering modular ‘ON’ RNA 50. Soukup GA, Breaker RR. 1999. Relationship between internucleotide switches using biological components. Nucleic Acids Res 41:10449– linkage geometry and the stability of RNA. RNA 5:1308–1325. http://dx 10461. http://dx.doi.org/10.1093/nar/gkt787. .doi.org/10.1017/S1355838299990891. 69. Robinson CJ, Vincent HA, Wu MC, Lowe PT, Dunstan MS, Leys D, 51. Regulski EE, Breaker RR. 2008. In-line probing analysis of ribo- Micklefield J. 2014. Modular riboswitch toolsets for synthetic genetic switches. Methods Mol Biol 419:53–67. http://dx.doi.org/10.1007/978-1 control in diverse bacterial species. J Am Chem Soc 136:10615–10624. -59745-033-1_4. http://dx.doi.org/10.1021/ja502873j. 52. Mandal M, Boese B, Barrick JE, Winkler WC, Breaker RR. 2003. 70. Porter EB, Polaski JT, Morck MM, Batey RT. 2017. Recurrent RNA Riboswitches control fundamental biochemical pathways in Bacillus motifs as scaffolds for genetically encodable small-molecule biosensors. subtilis and other bacteria. Cell 113:577–586. http://dx.doi.org/10.1016 Nat Chem Biol 13:295–301. http://dx.doi.org/10.1038/nchembio.2278. /S0092-8674(03)00391-X. 71. Suess B, Hanson S, Berens C, Fink B, Schroeder R, Hillen W. 2003. 53. Kulshina N, Baird NJ, Ferré-D’Amaré AR. 2009. Recognition of the Conditional gene expression by controlling translation with tetracycline- bacterial second messenger cyclic diguanylate by its cognate riboswitch. binding aptamers. Nucleic Acids Res 31:1853–1858. http://dx.doi.org Nat Struct Mol Biol 16:1212–1217. http://dx.doi.org/10.1038/nsmb.1701. /10.1093/nar/gkg285. 54. Stoddard CD, Montange RK, Hennelly SP, Rambo RP, Sanbonmatsu 72. Weigand JE, Sanchez M, Gunnesch EB, Zeiher S, Schroeder R, KY, Batey RT. 2010. Free state conformational sampling of the SAM-I Suess B. 2008. Screening for engineered neomycin riboswitches that con- riboswitch aptamer domain. Structure 18:787–797. http://dx.doi.org trol translation initiation. RNA 14:89–97. http://dx.doi.org/10.1261/rna /10.1016/j.str.2010.04.006. .772408.

10 ASMscience.org/MicrobiolSpectrum Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Small-Molecule-Binding Riboswitches

73. Groher F, Bofill-Bosch C, Schneider C, Braun J, Jager S, Geißler 92. Winkler WC, Nahvi A, Sudarsan N, Barrick JE, Breaker RR. K, Hamacher K, Suess B. 2018. Riboswitching with ciprofloxacin- 2003. An mRNA structure that controls gene expression by binding development and characterization of a novel RNA regulator. Nucleic S-adenosylmethionine. Nat Struct Biol 10:701–707. http://dx.doi.org Acids Res 46:2121–2132. http://dx.doi.org/10.1093/nar/gkx1319. /10.1038/nsb967. 74. Kim DS, Gusti V, Pillai SG, Gaur RK. 2005. An artificial riboswitch 93. Montange RK, Batey RT. 2006. Structure of the S-adenosylmethionine for controlling pre-mRNA splicing. RNA 11:1667–1677. http://dx.doi riboswitch regulatory mRNA element. Nature 441:1172–1175 http://dx .org/10.1261/rna.2162205. .doi.org/10.1038/nature04819. 75. Vogel M, Weigand JE, Kluge B, Grez M, Suess B. 2018. A small, 94. Nahvi A, Barrick JE, Breaker RR. 2004. Coenzyme B12 riboswitches portable RNA device for the control of exon skipping in mammalian cells. are widespread genetic control elements in prokaryotes. Nucleic Acids Res Nucleic Acids Res 46:e48. http://dx.doi.org/10.1093/nar/gky062. 32:143–150. http://dx.doi.org/10.1093/nar/gkh167. 76. Kötter P, Weigand JE, Meyer B, Entian KD, Suess B. 2009. A fast and 95. Cochrane JC, Lipchock SV, Strobel SA. 2007. Structural investigation fi ef cient translational control system for conditional expression of yeast of the GlmS ribozyme bound to its catalytic cofactor. Chem Biol 14:97– genes. Nucleic Acids Res 37:e120. http://dx.doi.org/10.1093/nar/gkp578. 105. http://dx.doi.org/10.1016/j.chembiol.2006.12.005. 77. Beisel CL, Chen YY, Culler SJ, Hoff KG, Smolke CD. 2011. Design of 96. Klein DJ, Wilkinson SR, Been MD, Ferré-D’Amaré AR. 2007. small molecule-responsive based on structural requirements Requirement of helix P2.2 and nucleotide G1 for positioning the cleavage – for Drosha processing. Nucleic Acids Res 39:2981 2994. http://dx.doi site and cofactor of the glmS ribozyme. J Mol Biol 373:178–189. http://dx .org/10.1093/nar/gkq954. .doi.org/10.1016/j.jmb.2007.07.062. fi 78. An CI, Trinh VB, Yokobayashi Y. 2006. Arti cial control of gene 97. Mandal M, Lee M, Barrick JE, Weinberg Z, Emilsson GM, Ruzzo WL, expression in mammalian cells by modulating RNA interference through Breaker RR. 2004. A glycine-dependent riboswitch that uses cooperative – aptamer-small molecule interaction. RNA 12:710 716. http://dx.doi.org binding to control gene expression. Science 306:275–279. http://dx.doi /10.1261/rna.2299306. .org/10.1126/science.1100829. fi 79. Ogawa A. 2011. Rational design of arti cial riboswitches based on 98. Huang L, Serganov A, Patel DJ. 2010. Structural insights into ligand ligand-dependent modulation of internal ribosome entry in wheat germ recognition by a sensing domain of the cooperative glycine riboswitch. – extract and their applications as label-free biosensors. RNA 17:478 488. Mol Cell 40:774–786. http://dx.doi.org/10.1016/j.molcel.2010.11.026. http://dx.doi.org/10.1261/rna.2433111. 99. Corbino KA, Barrick JE, Lim J, Welz R, Tucker BJ, Puskarz I, Mandal 80. Nomura Y, Zhou L, Miu A, Yokobayashi Y. 2013. Controlling M, Rudnick ND, Breaker RR. 2005. Evidence for a second class of mammalian gene expression by allosteric hepatitis delta virus ribozymes. S-adenosylmethionine riboswitches and other regulatory RNA motifs in – ACS Synth Biol 2:684 689. http://dx.doi.org/10.1021/sb400037a. alpha-proteobacteria. Genome Biol 6:R70. http://dx.doi.org/10.1186/gb 81. Felletti M, Stifel J, Wurmthaler LA, Geiger S, Hartig JS. 2016. Twister -2005-6-8-r70. fi ribozymes as highly versatile expression platforms for arti cial riboswitches. 100. Gilbert SD, Rambo RP, Van Tyne D, Batey RT. 2008. Structure of Nat Commun 7:12834. http://dx.doi.org/10.1038/ncomms12834. the SAM-II riboswitch bound to S-adenosylmethionine. Nat Struct Mol 82. Wieland M, Benz A, Klauser B, Hartig JS. 2009. Artificial ribozyme Biol 15:177–182. http://dx.doi.org/10.1038/nsmb.1371. switches containing natural riboswitch aptamer domains. Angew Chem 101. Fuchs RT, Grundy FJ, Henkin TM. 2006. The S(MK) box is a new – Int Ed Engl 48:2715 2718. http://dx.doi.org/10.1002/anie.200805311. SAM-binding RNA for translational regulation of SAM synthetase. Nat 83. Win MN, Smolke CD. 2008. Higher-order cellular information pro- Struct Mol Biol 13:226–233. http://dx.doi.org/10.1038/nsmb1059. – cessing with synthetic RNA devices. Science 322:456 460. http://dx.doi 102. Lu C, Smith AM, Fuchs RT, Ding F, Rajashankar K, Henkin TM, .org/10.1126/science.1160311. Ke A. 2008. Crystal structures of the SAM-III/S(MK) riboswitch reveal the 84. Beilstein K, Wittmann A, Grez M, Suess B. 2015. Conditional con- SAM-dependent translation inhibition mechanism. Nat Struct Mol Biol trol of mammalian gene expression by tetracycline-dependent hammer- 15:1076–1083. http://dx.doi.org/10.1038/nsmb.1494. – head ribozymes. ACS Synth Biol 4:526 534. http://dx.doi.org/10.1021 103. Park SY, Cromie MJ, Lee EJ, Groisman EA. 2010. A bacterial mRNA /sb500270h. leader that employs different mechanisms to sense disparate intracellular 85. Cressina E, Chen L, Moulin M, Leeper FJ, Abell C, Smith AG. 2011. signals. Cell 142:737–748. http://dx.doi.org/10.1016/j.cell.2010.07.046. fi Identi cation of novel ligands for thiamine pyrophosphate (TPP) riboswitches. 104. Cromie MJ, Shi Y, Latifi T, Groisman EA. 2006. An RNA sensor for – Biochem Soc Trans 39:652 657. http://dx.doi.org/10.1042/BST0390652. intracellular Mg2+. Cell 125:71–84. http://dx.doi.org/10.1016/j.cell.2006 86. Johnson JE Jr, Reyes FE, Polaski JT, Batey RT. 2012. B12 cofac- .01.043. – tors directly stabilize an mRNA regulatory switch. Nature 492:133 137. 105. Dann CE III, Wakeman CA, Sieling CL, Baker SC, Irnov I, Winkler http://dx.doi.org/10.1038/nature11607. WC. 2007. Structure and mechanism of a metal-sensing regulatory RNA. 87. Serganov A, Huang L, Patel DJ. 2009. Coenzyme recognition and gene Cell 130:878–892. http://dx.doi.org/10.1016/j.cell.2007.06.051. fl – regulation by a avin mononucleotide riboswitch. Nature 458:233 237. 106. Ramesh A, Wakeman CA, Winkler WC. 2011. Insights into http://dx.doi.org/10.1038/nature07642. metalloregulation by M-box riboswitch RNAs via structural analysis of 88. Vicens Q, Mondragón E, Batey RT. 2011. Molecular sensing by manganese-bound complexes. J Mol Biol 407:556–570. http://dx.doi.org the aptamer domain of the FMN riboswitch: a general model for ligand /10.1016/j.jmb.2011.01.049. – binding by conformational selection. Nucleic Acids Res 39:8586 8598. 107. Pikovskaya O, Polonskaia A, Patel DJ, Serganov A. 2011. Structural http://dx.doi.org/10.1093/nar/gkr565. principles of nucleoside selectivity in a 2′-deoxyguanosine riboswitch. Nat 89. Gilbert SD, Love CE, Edwards AL, Batey RT. 2007. Mutational Chem Biol 7:748–755. http://dx.doi.org/10.1038/nchembio.631. analysis of the aptamer domain. Biochemistry 46: 108. Kang M, Peterson R, Feigon J. 2009. Structural insights into ribo- – 13297 13309. http://dx.doi.org/10.1021/bi700410g. switch control of the biosynthesis of queuosine, a modified nucleotide 90. Garst AD, Héroux A, Rambo RP, Batey RT. 2008. Crystal structure found in the anticodon of tRNA. Mol Cell 33:784–790. http://dx.doi.org of the regulatory mRNA element. J Biol Chem 283: /10.1016/j.molcel.2009.02.019. – 22347 22351. http://dx.doi.org/10.1074/jbc.C800120200. 109. Klein DJ, Edwards TE, Ferré-D’Amaré AR. 2009. Cocrystal structure 91. Serganov A, Huang L, Patel DJ. 2008. Structural insights into amino of a class I preQ1 riboswitch reveals a pseudoknot recognizing an essential acid binding and gene control by a lysine riboswitch. Nature 455:1263– hypermodified nucleobase. Nat Struct Mol Biol 16:343–344. http://dx 1267. http://dx.doi.org/10.1038/nature07326. .doi.org/10.1038/nsmb.1563.

ASMscience.org/MicrobiolSpectrum 11 Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36 Lotz and Suess

110. Spitale RC, Torelli AT, Krucinska J, Bandarian V, Wedekind JE. 128. Ren A, Patel DJ. 2014. c-di-AMP binds the ydaO riboswitch in two 2009. The structural basis for recognition of the preQ0 metabolite by an pseudo-symmetry-related pockets. Nat Chem Biol 10:780–786. http://dx unusually small riboswitch aptamer domain. J Biol Chem 284:11012– .doi.org/10.1038/nchembio.1606. 11016 http://dx.doi.org/10.1074/jbc.C900024200. 129. McCown PJ, Liang JJ, Weinberg Z, Breaker RR. 2014. Structural, 111. Jenkins JL, Krucinska J, McCarty RM, Bandarian V, Wedekind JE. functional, and taxonomic diversity of three preQ1 riboswitch classes. Chem 2011. Comparison of a preQ1 riboswitch aptamer in metabolite-bound Biol 21:880–889. http://dx.doi.org/10.1016/j.chembiol.2014.05.015. and free states with implications for gene regulation. J Biol Chem 286: 130. Liberman JA, Suddala KC, Aytenfisu A, Chan D, Belashov IA, Salim 24626–24637. http://dx.doi.org/10.1074/jbc.M111.230375. M, Mathews DH, Spitale RC, Walter NG, Wedekind JE. 2015. Structural 112. Regulski EE, Moy RH, Weinberg Z, Barrick JE, Yao Z, Ruzzo WL, analysis of a class III preQ1 riboswitch reveals an aptamer distant from Breaker RR. 2008. A widespread riboswitch candidate that controls bac- a ribosome-binding site regulated by fast dynamics. Proc Natl Acad Sci terial genes involved in molybdenum cofactor and tungsten cofactor me- USA112:E3485–E3494. http://dx.doi.org/10.1073/pnas.1503955112. tabolism. Mol Microbiol 68:918–932. http://dx.doi.org/10.1111/j.1365 131. Dambach M, Sandoval M, Updegrove TB, Anantharaman V, -2958.2008.06208.x. Aravind L, Waters LS, Storz G. 2015. The ubiquitous yybP-ykoY ribo- 113. Wang JX, Lee ER, Morales DR, Lim J, Breaker RR. 2008. Ribo- switch is a manganese-responsive regulatory element. Mol Cell 57:1099– switches that sense S-adenosylhomocysteine and activate genes involved 1109. http://dx.doi.org/10.1016/j.molcel.2015.01.035. in coenzyme recycling. Mol Cell 29:691–702. http://dx.doi.org/10.1016 132. Price IR, Gaballa A, Ding F, Helmann JD, Ke A. 2015. Mn2+-sensing /j.molcel.2008.01.012. mechanisms of yybP-ykoY orphan riboswitches. Mol Cell 57:1110–1123. 114. Edwards AL, Reyes FE, Héroux A, Batey RT. 2010. Structural basis http://dx.doi.org/10.1016/j.molcel.2015.02.016. for recognition of S-adenosylhomocysteine by riboswitches. RNA 16: 133. Kellenberger CA, Wilson SC, Hickey SF, Gonzalez TL, Su Y, 2144–2155. http://dx.doi.org/10.1261/rna.2341610. Hallberg ZF, Brewer TF, Iavarone AT, Carlson HK, Hsieh YF, Hammond 115. Weinberg Z, Regulski EE, Hammond MC, Barrick JE, Yao Z, Ruzzo MC. 2015. GEMM-I riboswitches from Geobacter sense the bacterial WL, Breaker RR. 2008. The aptamer core of SAM-IV riboswitches mimics second messenger cyclic AMP-GMP. Proc Natl Acad Sci U S A 112:5383– the ligand-binding site of SAM-I riboswitches. RNA 14:822–828. http:// 5388. http://dx.doi.org/10.1073/pnas.1419328112. dx.doi.org/10.1261/rna.988608. 134. Nelson JW, Sudarsan N, Phillips GE, Stav S, Lünse CE, McCown PJ, 116. Sudarsan N, Lee ER, Weinberg Z, Moy RH, Kim JN, Link KH, Breaker RR. 2015. Control of bacterial exoelectrogenesis by c-AMP-GMP. Breaker RR. 2008. Riboswitches in eubacteria sense the second messenger Proc Natl Acad Sci U S A 112:5389–5394. http://dx.doi.org/10.1073 cyclic di-GMP. Science 321:411–413. http://dx.doi.org/10.1126/science /pnas.1419264112. .1159519. 135. Ren A, Wang XC, Kellenberger CA, Rajashankar KR, Jones RA, 117. Smith KD, Lipchock SV, Ames TD, Wang J, Breaker RR, Strobel Hammond MC, Patel DJ. 2015. Structural basis for molecular discrimi- SA. 2009. Structural basis of ligand binding by a c-di-GMP riboswitch. Nat nation by a 3′,3′-cGAMP sensing riboswitch. Cell Rep 11:1–12. http://dx Struct Mol Biol 16:1218–1223. http://dx.doi.org/10.1038/nsmb.1702. .doi.org/10.1016/j.celrep.2015.03.004. 118. Poiata E, Meyer MM, Ames TD, Breaker RR. 2009. A variant 136. Furukawa K, Ramesh A, Zhou Z, Weinberg Z, Vallery T, Winkler riboswitch aptamer class for S-adenosylmethionine common in marine WC, Breaker RR. 2015. Bacterial riboswitches cooperatively bind Ni2+ or bacteria. RNA 15:2046–2056. http://dx.doi.org/10.1261/rna.1824209. Co2+ ions and control expression of heavy metal transporters. Mol Cell 119. Lee ER, Baker JL, Weinberg Z, Sudarsan N, Breaker RR. 2010. An 57:1088–1098. http://dx.doi.org/10.1016/j.molcel.2015.02.009. allosteric self-splicing ribozyme triggered by a bacterial second messenger. 137. Kim PB, Nelson JW, Breaker RR. 2015. An ancient riboswitch class Science 329:845–848. http://dx.doi.org/10.1126/science.1190713. in bacteria regulates purine biosynthesis and one-carbon metabolism. Mol 120. Smith KD, Shanahan CA, Moore EL, Simon AC, Strobel SA. 2011. Cell 57:317–328. http://dx.doi.org/10.1016/j.molcel.2015.01.001. Structural basis of differential ligand recognition by two classes of bis- 138. Ren A, Rajashankar KR, Patel DJ. 2015. Global RNA fold and (3′-5′)-cyclic dimeric guanosine monophosphate-binding riboswitches. molecular recognition for a pfl riboswitch bound to ZMP, a master reg- Proc Natl Acad Sci U S A 108:7757–7762. http://dx.doi.org/10.1073/pnas ulator of one-carbon metabolism. Structure 23:1375–1381. http://dx.doi .1018857108. .org/10.1016/j.str.2015.05.016. 121. Ames TD, Rodionov DA, Weinberg Z, Breaker RR. 2010. A eubac- 139. Li S, Hwang XY, Stav S, Breaker RR. 2016. The yjdF riboswitch terial riboswitch class that senses the coenzyme tetrahydrofolate. Chem candidate regulates gene expression by binding diverse azaaromatic com- Biol 17:681–685. http://dx.doi.org/10.1016/j.chembiol.2010.05.020. pounds. RNA 22:530–541. http://dx.doi.org/10.1261/rna.054890.115. 122. Trausch JJ, Ceres P, Reyes FE, Batey RT. 2011. The structure of a 140. Nelson JW, Atilho RM, Sherlock ME, Stockbridge RB, Breaker RR. tetrahydrofolate-sensing riboswitch reveals two ligand binding sites in a 2017. Metabolism of free guanidine in bacteria is regulated by a wide- single aptamer. Structure 19:1413–1423. http://dx.doi.org/10.1016/j.str spread riboswitch class. Mol Cell 65:220–230. http://dx.doi.org/10.1016 .2011.06.019. /j.molcel.2016.11.019. 123. Ames TD, Breaker RR. 2011. Bacterial aptamers that selectively bind 141. Reiss CW, Xiong Y, Strobel SA. 2017. Structural basis for ligand glutamine. RNA Biol 8:82–89. http://dx.doi.org/10.4161/rna.8.1.13864. binding to the guanidine-I riboswitch. Structure 25:195–202. http://dx 124. Ren A, Xue Y, Peselis A, Serganov A, Al-Hashimi HM, Patel DJ. .doi.org/10.1016/j.str.2016.11.020. 2015. Structural and dynamic basis for low-affinity, high-selectivity 142. Sherlock ME, Malkowski SN, Breaker RR. 2017. Biochemical vali- binding of L-glutamine by the glutamine riboswitch. Cell Rep 13:1800– dation of a second guanidine riboswitch class in bacteria. Biochemistry 1813. http://dx.doi.org/10.1016/j.celrep.2015.10.062. 56:352–358. http://dx.doi.org/10.1021/acs.biochem.6b01270. 125. Baker JL, Sudarsan N, Weinberg Z, Roth A, Stockbridge RB, Breaker 143. Huang L, Wang J, Lilley DM. 2017. The structure of the guanidine-II RR. 2012. Widespread genetic switches and toxicity resistance proteins for riboswitch. Cell Chem Biol 24:695–702.e2. http://dx.doi.org/10.1016/j fluoride. Science 335:233–235. http://dx.doi.org/10.1126/science.1215063. .chembiol.2017.05.014. 126. Ren A, Rajashankar KR, Patel DJ. 2012. Fluoride ion encapsulation 144. Sherlock ME, Breaker RR. 2017. Biochemical validation of a third by Mg2+ ions and phosphates in a fluoride riboswitch. Nature 486:85–89. guanidine riboswitch class in bacteria. Biochemistry 56:359–363. http:// http://dx.doi.org/10.1038/nature11152. dx.doi.org/10.1021/acs.biochem.6b01271. 127. Nelson JW, Sudarsan N, Furukawa K, Weinberg Z, Wang JX, Breaker 145. Huang L, Wang J, Wilson TJ, Lilley DM. 2017. Structure of the RR. 2013. Riboswitches in eubacteria sense the second messenger c-di-AMP. guanidine III riboswitch. Cell Chem Biol 24:1407–1415.e2. http://dx.doi Nat Chem Biol 9:834–839. http://dx.doi.org/10.1038/nchembio.1363. .org/10.1016/j.chembiol.2017.08.021.

12 ASMscience.org/MicrobiolSpectrum Downloaded from www.asmscience.org by IP: 157.89.65.129 On: Wed, 08 Aug 2018 13:08:36