the purine riboswitch as a model system for exploring rna biology and chemistry

Biochimica et Biophysica Acta xxx (2014) xxx–xxx

BBAGRM-00677; No. of pages: 12; 4C: 2, 3, 4, 5, 8, 10

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r .com/ locate /bbagrm

Review

The purine riboswitch as a model system for exploring RNA biologyand chemistry☆

Ely B. Porter, Joan G. Marcano-Velázquez, Robert T. Batey ⁎Department of Chemistry and Biochemistry, 596 UCB, University of Colorado, Boulder, CO 80309-0596, USA

☆ This article is part of a Special Issue entitled: Riboswi⁎ Corresponding author. Tel.: +1 303 735 2159; fax: +

E-mail address: [email protected] (R.T. Batey

http://dx.doi.org/10.1016/j.bbagrm.2014.02.0141874-9399/© 2014 Elsevier B.V. All rights reserved.

Please cite this article as: E.B. Porter, et al., ThActa (2014), http://dx.doi.org/10.1016/j.bba

a b s t r a c t
a r t i c l e i n f o
Article history:Received 25 November 2013Received in revised form 17 February 2014Accepted 20 February 2014Available online xxxx

Keywords:RiboswitchPurineRNA structureCo-transcriptional foldingGene expression

Over the past decade the purine riboswitch, and in particular its nucleobase-binding aptamer domain, hasemerged as an importantmodel system for exploring various aspects of RNA structure and function. Its relativelysmall size, structural simplicity and readily observable activity enable application of awide variety of experimen-tal approaches towards the study of this RNA. These analyses have yielded important insights into smallmoleculerecognition, co-transcriptional folding and secondary structural switching, and conformational dynamics thatserve as a paradigm for other RNAs. In this article, the current state of understanding of the purine riboswitchfamily and how this growing knowledge base is starting to be exploited in the creation of novel RNA devicesare examined. This article is part of a Special Issue entitled: Riboswitches.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Over the past decade, structural and biophysical studies of RNA havebeen heavily influenced by riboswitches, and in particular, purineriboswitches. These regulatory RNA elements are most prevalent inthe leader sequences of bacterial mRNAs, controlling expression in acis-fashion (reviewed in [1]). This activity is imparted through two func-tional domains: a smallmolecule binding aptamer (or receptor) domainand a regulatory component (or expression platform) containing astructural switch that most often acts at the level of transcription ortranslation. Structural and biophysical studies of riboswitches have al-most exclusively focused upon the aptamer domain as this domainhas both complex tertiary architecture amenable to structure determi-nation and an observable activity—ligand binding. As a consequence,while three-dimensional architectures of almost every major family ofriboswitch have been elucidated, only two structures (the SAM-III (orSMK) [2] and hydroxocobalamin riboswitch [3]) have been shown di-rectly and experimentally to encompass all of the sequence elementsnecessary and sufficient to impart both the ligand binding and gene reg-ulatory activities. While several riboswitch aptamer domains such asSAM-I [4], TPP [5,6], and preQ1-I [7–9] have been extensively studiedby a number of approaches, the aptamer domain of members of the pu-rine riboswitch family [10,11] have been adopted as a significant modelsystem for studying these RNAs.

tches.1 303 492 5894.).

e purine riboswitch as a modgrm.2014.02.014

The purine riboswitch was originally discovered as a conserved reg-ulatory feature in leader sequences of mRNAs associated with purinemetabolism [12]. This study recognized two mutually exclusive stem–

loop structures in the mRNA leader sequence, a Rho-independent tran-scriptional terminator and an antiterminator, secondary structural fea-tures of the mRNA known to be associated with transcriptionalattenuation [13]. In addition, repression of transcription was effectedby lowmolecularweight compounds hypoxanthine, guanine andweak-ly, xanthine. However, a proposed regulatory protein mediating thisprocess, such as those acting on the trp and pyr operons, was not iden-tified. Similar findings with leader sequence elements associated withoperons related to S-adenosylmethionine [14], thiamine pyrophosphate(TPP) [15], riboflavin [16] and adenosylcobalamin [17] metabolism ledto the very prescient speculation that these mRNAs directly bind theirsmall molecule effectors [18]. Only after the discovery of several otherconserved RNA elements that directly bind to small molecule metabo-lites [19–23] was it established that some purine leader sequences in-teract with either guanine or adenine [24,25]. Shortly after thisdiscovery, the Batey laboratory was the first to report the structure ofa riboswitch, the Bacillus subtilis xpt–pbuX guanine-responsive aptamerdomain, revealing core principles of RNA-small molecule recognitionand regulation of gene expression by riboswitches [10]. Later, a thirdclass of riboswitches within the purine riboswitch family was discov-ered to be regulated by 2′-deoxyguanosine, but this RNA is currentlyonly found in a single organism, Mesoplasma florum [26].

There are several reasonswhy the purine riboswitch has emerged asa favored model system. First, and likely foremost, it is a structurallysimple “complex” RNA—that is a multi-helix structure with helicalpacking. This architecture, described as a three-way junction supported

el system for exploring RNA biology and chemistry, Biochim. Biophys.

http://dx.doi.org/10.1016/j.bbagrm.2014.02.014

mailto:[email protected]


http://www.sciencedirect.com/science/journal/18749399


2 E.B. Porter et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

by a distal tertiary interaction (see Section 2), is a highly recurrenttheme in RNA biology [27]. In addition to the purine riboswitch aptamerthis structural theme is employed to create a functional center in severalother riboswitch aptamer domains, the hammerhead ribozyme, signalrecognition particle and various substructures of the ribosomal RNA.Further, of the three classes of three-way junctions, the purineriboswitch is a member of the most common class (class C) [28]. Sec-ondly, this RNA has several activities—namely ligand binding and generegulation—that can be monitored by standard molecular,biochemical, and biophysical techniques. A significant advantage is ad-enine binding variants that also recognize 2-aminopurine (2AP), en-abling fluorescence approaches to be used that do not require labelingof the RNA or ligand. Third, RNAs derived from purine riboswitchaptamer domains are remarkably well behaved. While many RNAstend to misfold, multimerize, or readily degrade, the purine bindingaptamer has proven itself easy to work with. In part, this is likely dueto the strong selective pressures on rapid and efficient folding presentinside the cell as riboswitches may have only a brief timeframe inwhich to influence the expression machinery (see Section 6). Finally,the purine riboswitch aptamer at ~60 nucleotides in length is one ofthe smallest RNAs that embodies the above features. The small sizemakes NMR spectroscopy tractable as well as facilitates the facile incor-poration of modifications, such as fluorophores, into the RNA. In this re-view, we discuss how the purine riboswitch has been used as a modelsystem to study aspects of RNA chemistry and biology.

2. Purine nucleobase recognition

Currently, the aptamer domain structure of fourmembers of the pu-rine riboswitch family has been solved by X-ray crystallography:B. subtilis xpt–pbuX guanine-responsive [10,11], Vibrio vulnificus addadenine-responsive [11], B. subtilis pbuE adenine-responsive [29], andM. florum 1A 2′-deoxyguanosine-responsive [30]. The aptamer domainsof these four riboswitches share a common three-dimensional

Fig. 1. Thepurine riboswitch consensus and binding architecture. (A) A simplified consensus seqfrom the Rfam database [33]. Red nucleotides indicate at least 90% conservation. Colored backg(B) Global architecture of a representative member of the purine family (B. subtilis xpt–pbuX) bpattern of the two residues that confer ligand specificity (xpt:hypoxanthine U51 and C74) and

Please cite this article as: E.B. Porter, et al., The purine riboswitch as a modActa (2014), http://dx.doi.org/10.1016/j.bbagrm.2014.02.014

architecture and ligand binding pocket, which are highly conservedthroughout the family [24,26,31]. The aptamer domain consists ofthreeWatson–Crick paired regions (P1–P3; Fig. 1A, B) organized arounda central three-way junction. Sequence conservation within theaptamer is localized to the two terminal loops (L2 and L3) that form apseudoknot interaction and the three joining regions (J1/2, J2/3 andJ3/1) that comprise the junction. Within the junction is a set of nearlyinvariant nucleotides (highly conserved nucleotides across the purinefamily [32,33] are highlighted in red in Fig. 1A) organizing three criticalbases for hydrogen bonding interactionswith the ligand forming essen-tially a base quartet (nucleotides 47, 51, and 74 in the xpt numberingscheme; for simplicity, features of all members of the purine family inthis review will be referenced using this numbering scheme). Thesebases, along with the 2′-hydroxyl group of U22, create the network ofhydrogen bonds that fully recognize the ligand (Fig. 1C). Adenine-responsive riboswitches discriminate against guanine primarilythrough a single change in the pocket (U74 for C74), enablingWatson–Crick pairing to this nucleobase. While in principle U74 couldpair to guanine through a wobble pair, the architecture of the junctionappears to prevent shifting of nucleotide 74 towards the major groove[34].

Recognition of 2′-deoxyguanosine, on the other hand, is achievedprimarily through the identity of the nucleotide 51 being cytosine(Fig. 1A) [30,35]. In the bound state, C51 shifts towards 74 relative toU51 in the guanine-bound aptamer, sterically accommodating the 2′-deoxyribose sugar. Accordingly, nucleotide 47disengages from its inter-actions with residue 51 and reorients itself out towards the solvent.Thus, it appears that identities of just twonucleotides, 51 and 74, governdiscrimination between different purine nucleobases and nucleosides[35]. In principle, the C51/U74 combination would enable recognitionof 2′-deoxyadenosine, although this sequence has not yet been ob-served in biology [26]. Additionally, the M. florum 1A riboswitch has a~100-fold preference for 2′-deoxyguanosine over guanosine [26]. Com-parison of the wild type aptamer bound to both compounds revealed

uence projectedupon the secondary structure of thepurine family of riboswitches, derivedrounds represent paired regions corresponding to the overall three-dimensional structure.ound to hypoxanthine (HX). Structure is taken from PDB ID: 1U8D. (C) Hydrogen bondingthe critical organizing nucleotide U47.



3E.B. Porter et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx

that the ribose sugar of guanosine adopts the C3′-endo conformationthat prevents hydrogen bonding of the 3′-hydroxyl group with C48(C56 in M. florum numbering) [30]. This causes C48 to be flipped outtowards the solvent, whichwas speculated to destabilize the conforma-tion of J2/3, leading to lower affinity for guanosine.

Other riboswitches use similar pyrimidine-rich binding sites to rec-ognize purine nucleobases. For example, the three-way junction(3WJ) site of the THF riboswitch [36] and the preQ1-II riboswitch [37]recognizes guanine through a U-purine-C base triple (Fig. 2A, B). Thesecond site in the THF riboswitch, at the pseudoknot (PK), has a slightlydifferent arrangement of pyrimidine residues around the purine base,but still uses a uridine base to recognize the N9/N3 face (note that theN9/N3 face corresponds to the “sugar edge” of purine nucleotides)(Fig. 2C) [36]. The only deviation from this theme is found in thepreQ1-I riboswitch where the N9/N3 face of the guanine base interactswith the Watson–Crick face of the base of an adenosine residue(Fig. 2D) [7–9]. Similarly, the in vitro selected theophylline aptameruses two pyrimidine residues to hydrogen bond to the Hoogsteen andN3/N9 edges of its ligand [38]. Thus, for most aptamers recognizingpurine nucleobases, a recognition strategy involving two pyrimidineresidues appears to be the easiest solution. This simple scheme,however, is not reflected in the binding of purine nucleobases as partof more complex molecules such as S-adenosylmethionine, cyclic-di-GMP or adenosylcobalamin by riboswitches as well as GMP recognitionby group I introns. In these cases, recognition of the nucleobase of the li-gand is idiosyncratic, more typically using hydrogen bonding interac-tions with purines in the RNA [39–41].

The purine riboswitches also bind a variety of non-natural purine[24,31,34,42–44] and pyrimidine analogs [45,46]. Many of these analogsrepresent trivial solutions to recognition in that they bindwith the samehydrogen bonding configuration as the natural effectors but with no al-teration in the conformation of the RNA. For example, 2-aminopurine(2AP) binds to the adenine riboswitch and the (C74U)xpt–pbuXriboswitch in the same configuration as adenine, enabling it to bewide-ly exploited as a means of investigating the ligand binding properties ofpurine riboswitches due to its spectral properties [43,47–49].

2AP recognition by purine riboswitches revealed an interesting as-pect of effector discrimination by these RNAs. Several compounds

Fig. 2. Comparing the modes of recognition between various purine binding pockets. A commooften with a uracil base recognizing the N3/N9 face of the ligand. The binding pocket of the (Aidentical, with recognition of the ligand achieved by the same spatial arrangement. Again in thehydrogen bonds are mediated by two ancillary pyrimidines juxtaposed to the Watson–Crick fa3FU2) where the binding pocket is still populated by several pyrimidines, yet recognition is pa


with substitutions at the purine 6-position, such as 2-AP (deletion ofN6), 6-chloroguanine (substitution of the guanine O6 with a chlorineatom), and 6-O-methylguanine (methylation of the guanine O6), bindguanine and adenine riboswitches with nearly the same affinity. Thisis in stark contrast to the nearly 10,000-fold specificity each riboswitchhas for its associated purine (guanine vs. adenine) [24,25]. A compari-son of the structures of the wild type and C74U xpt aptamers bound to2AP, revealed that cytosine 74 is capable of shifting to theminor grooveto re-establish a two-hydrogen bonding interactionwith the ligand [34].Presumably, uridine 74 in adenine riboswitches cannot shift towardsthe major groove, which would enable a G·U wobble pair to form anddiminish the discrimination of this riboswitch for adenine over guanine.These observations highlight that in searches for compounds that bindpurine riboswitches as potential antimicrobial agents [50], the abilityof nucleotides 51 and 74 to reposition themselves within the ligandbinding pocket to establish alternative hydrogen bonding patternsmust be considered.

3. RNA structural nuances yield functional differences

Global organization of the purine riboswitch family is governed byinteractions between L2 and L3 serving to pack P2 against P3 (Fig. 1B).Under near physiological magnesium ion concentrations (0.5–1 mM),this interaction provides ~4.1 kcal/mol to the ligand binding energy inthe xpt/pbuX riboswitch [51] and at least 2.9 kcal/mol in the pbuE ade-nine riboswitch [52]. Formation of this tertiary interaction is criticalfor high affinity recognition of ligands, but low affinity binding can beachieved in its absence [51–53]. At the core of this interaction withinall members of the purine clan is the formation of two invariant G–CWatson–Crick pairs between L2 and L3 (“pseudoknot”, Fig. 1A). Beyondthis core element, the guanine, adenine, and 2′-deoxyguanosineriboswitches all have differences contributing to their effector bindingproperties.

The guanine riboswitches, as exemplified by the B. subtilis xpt–pbuXriboswitch, contain an extremely stable L2–L3 interaction capable offorming in the absence of either magnesium ion or ligand. The overallarchitecture of this tertiary interaction is a set of four non-canonicalbase pairs that scaffold two essential Watson–Crick G–C pairs (Fig. 3A,

n theme of pyrimidine-rich binding pockets is observed in most purine-binding aptamers,) THF three-way junction (PDB ID: 4LVW) and (B) class II preQ1 (PDB ID: 4JF2) is nearly(C) THF pseudoknot site, guanine's N3/N9 face is paired to a uracil base, while additionalce. The only major departure from this trend is the (D) class I preQ1 riboswitch (PDB ID:rtially achieved by an adenine base structured via auxiliary hydrogen bonding to a uracil.



Fig. 3. Spatial arrangement of the loop–loop tertiary interaction in various purine riboswitch aptamers. (A, B) The xpt–pbuX guanine sensing riboswitch contains a stable L2–L3 interactionwhere four non-canonical base pair interactions (yellow)provide the context for two essential and invariantWatson–Crick G–C pairs (orange) (PDB ID: 1U8D). These loops are both closedbyWatson–Crick base pairs (green). (C, D) The pbuE riboswitch contains a highly similar tertiary interaction as xpt (PDB ID: 3IVN) but differs in that both loops are closed by non-canonicalbase pairs. (E, F) The 2′-deoxyguanosine aptamer deviates substantially in the base identities surrounding the G–C pairs containing a deletion of three nucleotides in L3 (PDB ID: 3SKI). Inall panels, hydrogen bonds are represented as gray dashes, and select oxygen and nitrogen atoms are red and blue, respectively.


orange). The G38–C60 Watson–Crick pair interacts with a reverseWatson–Crick/Hoogsteen A33–A66 pair in its minor groove, while theG37–C61 pair interacts with a reverse Watson–Crick/Hoogsteen U34–A65 pair. It is interesting to note that the packing of the two non-canonical base pairs into the minor groove of the G–C pairs is reminis-cent of the type-I/type-II A-minor triples that define the interaction ofGAAA tetraloops into the minor groove of A-form helices [54,55]. Distalto these two base quartets are two other pairs: an A35·A63 pair and aside-by-side G62·U63 pair. These pairs make significantly weaker ener-getic contributions to ligand binding, as reflected in their lower degreeof phylogenetic conservation [53]. smFRET [56], NMR [57], and in-linechemical probing studies [24] all show that the L2–L3 tertiary interac-tion is fully formed and stable in the absence of guanine and moderate(1 mM) magnesium ion concentrations. Even in the absence of Mg2+,the “docked” conformation is significantly sampled, indicating thatmagnesium is not essential for the L2–L3 interaction [56]. Therefore, itis likely that this tertiary structure is tightly formed under physiologicalconditions such that it promotes preorganization of the binding pocketto enable rapid and efficient ligand binding [58]. Observations by NMRof ligand binding in the absence of magnesium reinforce this observa-tion [11,59], although the interaction is significantly weakened in theabsence of divalent ions [10]. Strikingly, the L2–L3 interaction is ob-served to form outside the context of the three-way junction or helix1, further emphasizing its intrinsic stability [57].

A number of adenine riboswitches contain a less stable L2–L3 inter-action as compared to the guanine riboswitches. While base specific in-teractions between L2 and L3 are nearly identical [11], the closing basepairs differ [11,31]. In the vast majority of guanine riboswitches, theclosing pair of L2 and L3 is a Watson–Crick base pair (Fig. 3B), whereasin the B. subtilis pbuE adenine riboswitch, L2 is closed by a U·U pair andL3 by anA·Apair (L3 of the V. vulnificus add adenine riboswitch is closedby a G–C Watson–Crick pair) (Fig. 3C, D). In the pbuE variant, the ab-sence of magnesium does not promote even transient sampling of thedocked configuration [52], which is supported by NMR studies [60]. In-stead, at least 50 μMMg2+ is required to promote the docked state [61].Differences in the stability of the L2–L3 interaction may have conse-quences for the regulatory activity of the riboswitch because the tertiary


interaction directly influences the conformational ensemble of the un-bound junction. Destabilization of this interaction would result in in-creased disorder and potentially decrease the rate of ligand binding(kon) and thereby increasing the concentration of ligand required toelicit a regulatory response (see Section 6) [60].

Finally, the class 1A M. florum 2′-deoxyguanosine riboswitch con-tains a substantial variation in the nature of the L2–L3 interaction.While L2 possesses the same size and pattern of sequence conservationas the rest of the purine riboswitch family, L3 contains a deletion of themiddle three nucleotides (Fig. 3E, F). While the two coreWatson–Crickpairs are maintained, only the equivalent of G37–C69 is engaged byother nucleotides. Distal to these pairs, a single A·A pair between theloops further reinforces the tertiary interaction. Despite the significantdifference in tertiary architecture, these novel interactions contributelittle to the specificity for 2′-dG over guanine [30,35]. Instead, sequencedifferences in P2 adjacent to the three-way junction confer specificity[30,35].

Within P2 and proximal to the three-way junction is a lightly con-served yet important component called the “P2 tune box” [48]. Thissequence element was found through a structure-based sequencealignment of guanine/adenine riboswitches in which only the subsetof sequences alignablewithout insertions and/or deletionswere consid-ered. This high-quality alignment yields two sets of nucleotides show-ing statistically significant co-variation despite being non-interacting:66/68 in L3 and 24/25 at the interface of J1/2 and P2. Genetic and bio-chemical analyses revealed a covariation pattern between nucleotide24,which stacks between nucleotides 72 and 73 at the P3–J3/1 interfaceand the first two base pairs in P2. Within the purine family there is amarked tendency for the first two base pairs of P2 to be non-canonicalpairs. Interestingly, this is a crucial element in the 2′-deoxyguanineaptamer for achieving high-affinity and specific binding of its cognate li-gand—more important even than the variant L2–L3 interaction [35]. Thestructure of this aptamer revealed that nucleotide 24 instead of stackingwith P3, forms a triple base pair with the first base pair in P2 [30]. De-tailed affinity and kinetic measurements of a series of P2 tune box se-quences, both natural and non-natural, reveal that this region has asignificant impact upon the association and dissociation rates of ligand



Fig. 4. Coordination of metal ions along the purine riboswitch. (A) Magnesium ion (ma-genta) present in the crystal structure of the add riboswitch from B. subtilis (PDB ID:1Y26) that is adjacent to the binding pocket. (B) Two cobalt hexammines observed inthe xpt riboswitch structure (PDB ID: 1U8D) reveal the interaction of diffuse ions alongthe riboswitch bridging a region of the structure where the backbone comes into veryclose proximity to itself (distances between three non-bridging phosphate oxygensshown).


binding [48]. Strikingly, while a single point mutation in this box is le-thal to the in vivo regulatory activity of the riboswitch, a compensatorymutation to non-interacting nucleotides (i.e., to nucleotides not in-volved in base pairing) rescues full wild type activity. Like the distal L2sequence variation affecting the stability of the tertiary interaction, itis hypothesized that the P2 tune box also influences the conformationalensemble of the free state to tune the regulatory activity of theriboswitch to the needs of its associated transcriptional unit.

Another region with clear patterns of conservation exists within theP1 helix that has yet to be investigated. The crystal structure of the gua-nine and adenine riboswitches reveals that the two base pairs proximalto the ligand binding core (U20–A76 and A21–U75) are involved inligand-dependent base triples with nucleotides in J2/3 (49 and 50, re-spectively), and that the identity of these nucleotides is almost invariant(Fig. 1A). Beyond that, the next three base pairs show a significantpreference for the orientation of the purine–pyrimidine pair (for exam-ple, the pair below U20–A76 shows a significant tendency for a purineon the 5′-side of the helix). Most likely, these preferences reflect aneed to optimize the stability of P1 relative to the competing down-stream helix. However, like the rest of the expression platform (seeSection 5), the role of specific sequence elements or conservation pat-terns in the P1 helix remains poorly understood. This is somewhat sur-prising in light of the central role that this helix plays in communicatingthebinding state of the aptamer to thedownstreamelements that directthe expression machinery.

4. The role of metal ions

Like almost all RNAs, cations have a strong influence upon the struc-ture and activity of the purine riboswitch (an excellent overview of therole of cations in RNA is given by Williams and colleagues [62]). Multi-valent cations interact with RNA in three distinct ways [63], the firstbeing ion specific sites playing specific and critical roles in catalysis orbinding. Examples include the TPP [5,6,64] and flavin mononucleotide(FMN) [65,66] aptamer domains in which magnesium mediates inter-actions between phosphate groups of the ligand and RNA. Quite re-markably, the recent structure of the fluoride riboswitch has revealedhow specific magnesium ions completely mediate the halide–RNA in-teraction [67]. In addition, the magnesium and glycine riboswitchaptamers use specifically boundmagnesium cations to stabilize RNA ar-chitecture [68–70]. Invariably, these ions form inner sphere contactswith either the RNA or the ligand. Second are divalent ions associatingwith RNA non-specifically in regions of high electronegative surface po-tential that may or may not form inner sphere contacts with the RNA.Several examples of such ions are found in the purine riboswitch, suchas a magnesium ion that sits adjacent to the ligand binding site in theadd-adenine structure [11] (Fig. 4A) and cobalt hexamine ions foundalong P2 and P3 of the xpt-hypoxanthine structure where phosphategroups come into close proximity (Fig. 4B). These ions are often ob-served in X-ray structures, but different monovalent and/or divalentions can occupy these sites [71]. Finally, “diffuse” ions compose an at-mosphere of nonwell-localized ions around the RNA and their behavioris dictated by long-range electrostatics. From a series of crystallographicand NMR studies of representative purine aptamers, a number of high-occupancy metal ion binding sites dispersed throughout the structurehave been observed [10,11,72], but none of these appear to be essentialfor ligand recognition. Instead, the role of divalent metal ions is to in-crease the affinity of the aptamer for effector by stabilizing global RNAarchitecture. Thus, for the purine riboswitch, cations appear to play anon-specific role promoting folding and ligand binding.

More recently, Leipply and Draper developed a quantitative modeldescribing the effect ofmagnesiumon folding of theV. vulnificus add ad-enine riboswitch [73]. Structural characterization of this RNA by smallangle X-ray scattering (SAXS) and hydroxyl radical probing enabledfour states of the RNA to be defined: (1) an unbound and extendedform of the aptamer in which L2 and L3 do not interact, (2) a ligand


bound form with the L2 and L3 interaction unformed, (3) an unboundform displaying the L2–L3 tertiary interaction and (4) the bound andfolded state. These four states are connected in a thermodynamic cycleby three free energy values: (1) formation of the L2–L3 interaction(ΔGdock), (2) binding of ligand (ΔGLBP) and (3) a coupling factor linkingthe two processes (ΔGw). The magnesium dependence of each of theseenergies yields information about the mechanism(s) by which divalentcations act on the free energy folding landscape of the RNA. For the ad-enine riboswitch, magnesium appears to act on the RNA in two wayswith nearly equal effectiveness: promoting formation of the L2–L3 in-teraction and stabilizing the ligand binding pocket. Conversely, couplingof the tertiary interaction and the ligand binding pocket is only weaklymagnesium dependent, indicating that their linkage is largely a struc-tural feature of the aptamer [73].

Monovalent cations also influence the stability of the tertiary archi-tecture of the purine riboswitch aptamer domain. In the absence ofmagnesium, the thermal stability of the V. vulnificus add–DAP complexis strongly influenced by the identity of the monovalent cation [74]. Asthe radius of the monovalent cation increases down the group I series(Li+, Na+, K+, Rb+, and Cs+), stability of the complex decreases by al-most 3 kcal/mol. This effect is rationalized as a result of high chargedensity in the RNA and the access of small ions to sterically restricted re-gions of high electronegative potential. This effect is counter to thehigher desolvation energies of smaller monovalents such that innersphere ion–RNA interactions are more costly [75,76]. The lack of a spe-cificmonovalent effect, such as that observed for GAAA-tetraloop recep-tor for K+ [74,77], indicates that monovalent cations interact with




purine riboswitch aptamer domains as diffuse ions. In support of thistrend, a soak of the M. florum 2′-deoxyguanosine riboswitch with10 mM CsCl reveals three unique monovalent binding sites; two sittingin themajor groove of P2 and a third adjacent to the 2′-hydroxyl of C31and non-bridging phosphate oxygens of A32 (equivalent to U22, A23 inthe xptnumbering system) [30]. None of these sites suggest a significantnumber of direct RNA–ion interactions, but rather generally reflectregions of high negative potential.

5. Beyond the aptamer domain: the nebulous expression platform

While the aptamer domain, defined as the minimal RNA sequencenecessary to achieve high affinity binding of the cognate effector ligand,is structurally well characterized, the structural features of the expres-sion platform are largely unexplored. A serious limitation in the analysisof this region is that most alignments of riboswitches, particularly thosefound in Rfam [32,33] generally exclude features other than the easilyidentifiable and alignable aptamer domain. Structural elements in theexpression platform, such as terminator/antiterminator or sequester/antisequester hairpins, are proposed based upon computational sec-ondary structural prediction algorithms. While these programs aregood at predicting relatively strong elements of secondary structure,such as Rho-independent terminators, the presence ofweaker elementsthat may nonetheless influence the regulatory properties of the RNA re-mains largely unknown. Additionally, transcriptional termination ele-ments can vary significantly in their structure impeding identificationof consensus elements and conserved motifs [78].

Studies of other classes of riboswitches illustrate the need to consid-er structural elements beyond the aptamer and the generally obviousalternative secondary structural switch in the expression platform. Inthe expression platform of the B. subtilis lysC lysine riboswitch, there isa small hairpin structure, P6, that lies between the P1 helix and theterminator element that does not appear to be directly involved in thesecondary structural switch—it is present in the antiterminator and ter-minator states [79]. The presence of this stem–loop, which would becontiguous with the antiterminator helix, is important for efficientlysine-dependent regulation. Deletion of this hairpin without affectingthe antiterminator/terminator structures causes the riboswitch to beconstitutively “OFF”, whereas stabilization of this element prohibitsformation of the terminator in an in vitro single-turnover transcriptionassay. Presumably, this hairpin facilitates nucleation of the antiterminatorhelix at the expense of P1, preventing Rho-independent termination.While this element is not conserved in lysine riboswitches, these resultssuggest that sequence and structure in the expression platform that arenot directly involved in alternative structure formationmay be importantfor efficient function [79].

A second instructive example of how inconspicuous structural fea-tures adjacent to the aptamer domain can profoundly influence activityis found in the glycine riboswitch. Originally, this riboswitch family wasdefined by two structurally similar aptamer domains, each binding asingle glycine molecule with strong cooperativity [80]. However, justoutside of the presumed aptamer domain exists a recently discoveredkink-turn motif connecting the two aptamers [81,82]. This motif andthe associated P0 helix promote interdomain association in the absenceof glycine, abrogating cooperative ligand binding under physiologicalconditions [82,83]. These studies highlight the need to investigate thesequence and structure of expression platforms to fully define the rolethe expression platformplays in gene regulation by purine riboswitches.

6. Mechanisms: a tale of two adenine riboswitches

In bacteria, riboswitches typically control gene expression by tran-scriptional and/or translational attenuation, although a recent studyhas uncovered further evidence for widespread regulation of gene ex-pression via regulation of antisense RNAs [84]. Also, some riboswitcheshave been identified that require the Rho termination factor as part of


the regulationmechanism [85]. Regulatory control at the transcriptionallevel has a significant consequence: the riboswitch has only a shorttimeframe in which to influence RNA polymerase before it escapes be-yond the intrinsic terminator. Thus, the riboswitchmust do three thingsrapidly and efficiently: (1) acquire secondary and tertiary structure inthe aptamer domain, (2) survey the cellular environment for the cog-nate ligand, and (3) direct alternative secondary structure formationbased on the status of the aptamer domain. An early study of theB. subtilis pbuE adenine riboswitch revealed that adenine binds to theaptamer with slow association and dissociation kinetics [49]. A theoret-ical consideration of the timeframe required to fully saturate theaptamer at a given intracellular ligand concentration yielded the conclu-sion that the timeframe of equilibration of the aptamer is substantiallylonger than what is required for the polymerase to transcribe the ex-pression platform through the terminator. Thus, it was proposed thatthere is a subset of riboswitches under “kinetic control”. The hallmarkof kinetic control is that concentrations of effector required to elicit ahalf maximal regulatory response (referred to as the EC50 for in vivoand T50 for in vitro) are greater than the concentrations of ligand re-quired to half-saturate the aptamer under full equilibrium (referred toas the KD). Presumably, riboswitches that control gene expression atthe translational level do not have a similar temporal constraint, andthus potentially could be under “thermodynamic control” in whichthe EC50 or T50 is equivalent to the aptamer's KD.

To understand this feature of riboswitches, the Lafontaine and Masségroups performed a comparative study of the B. subtilis pbuE adenineriboswitch (transcriptional regulation) and the V. vulnificus add adenineriboswitch (translational regulation) [86]. It had been previouslyestablished that the full length pbuE riboswitch encompassing boththe aptamer domain and expression platform cannot efficiently bindadenine, indicating that it is thermodynamically locked into a statewhere the terminator stem–loop has formed at the expense of theaptamer [52,86,87]. This implies that the RNA does not act as a revers-ible switch, but rather as a “fuse” whose ligand-directed folding mustoccur during the transcription process [88].

The regulatory properties of the pbuE riboswitch are strongly influ-enced by transcription conditions. First, in a single-turnover transcriptionassay the T50 showed a significant dependence upon the concentration ofNTPs [86], which influences the rate of transcription [89]. As the concen-tration of NTPs was raised from 20 to 150 μM, the T50 increases from 0.2to 1.3 μM for 2,6-diaminopurine (an adenine analog). Similar resultshave been observed for the B. subtilis ribD FMN [88] and lysC lysine[90] riboswitches. Second, a long-lived transcriptional pause locateddownstream of the pbuE adenine riboswitch aptamer domain was ob-served at a stretch of uridine residues at positions 114–117 [86]. Uridinetracts are well-known pause sites for bacterial RNAPs, and a similarfinding was again observed in the FMN riboswitch [49]. Presumably,the pause stalls the polymerase affording the aptamer more time to in-terrogate the cellular environment and reach equilibrium with respectto effector concentration. Third, NusA, a promiscuous acting transcrip-tion factor that generally affects termination, lowers the T50 by decreas-ing the rate of transcription. Finally, all of the observed T50 values aregreater than that for DAP binding the pbuE riboswitch (KD ~ 25 nM)[86]. Together, these data indicate that the riboswitch only acts in thecontext of transcription and is modulated by factors affecting the rateof RNA synthesis by means of altering the residence time of the poly-merase at the riboswitch. While these in vitro data strongly suggestthat pbuE is under kinetic control, analogous experiments need to beperformed in vivo to firmly establish that in cellular conditions, thesame properties are observed.

Similar experiments on the V. vulnificus add riboswitch indicate thatit exerts translational control in a thermodynamic regime. In contrast tothe pbuE riboswitch, the full length add riboswitch binds adenine withonly a slight reduction in affinity as compared to the aptamer domainalone [86,87]. This suggests that the alternative P1 stem and sequesterhairpin reversibly exchange between the two states. This was




confirmed by chemical probing of the native and several mutantsequences that stabilized either the “ON” or “OFF” state in the pres-ence and absence of adenine as well as by single molecule forceextension spectroscopy [86,91]. Further, this riboswitch upregulatesβ-galactosidase expression in either a coupled or uncoupled in vitrotranslation system, again indicating the reversibility of the switch [86].However, in vivo, the riboswitch requires ~500 μMadenine in the extra-cellular media to elicit a half-maximal regulatory response, substantial-ly higher than the KD (~500 nM). While this might be interpreted askinetic control in vivo, the authors point out that the cellular uptakeof adenine may be very poor or the metabolic turnover high, such thatthe intracellular concentration of adenine is substantially lower thanthat in the medium. Further, the pioneer round of translation is physi-cally coupled to transcription in vivo via NusG [92], which also stimu-lates Rho-dependent transcriptional termination. Thus, translationalriboswitches may face temporal constraints via a competition betweenthe rate of ribosome association and Rho-dependent termination. It re-mains to be determinedwhether the thermodynamic control of the addriboswitch in vitro also occurs in the cell. While it may be a temptingspeculation to associate translational regulation with thermodynamiccontrol [86], this is still an open question.

Another excellent study addressing the question of thermodynamicversus kinetic control of riboswitches investigated the Fusobacteriumnucleatum preQ1-1 riboswitch [93]. A critical feature of this riboswitchis that the expression platform is “bistable” in which the antiterminatorand terminator hairpins are in reversible equilibrium, where ligandbinding to the aptamer significantly shifts the equilibrium towards theterminator state. This nearly unique property enabled a detailed charac-terization of the ligand-dependent switching properties of the fulllength riboswitch using both NMR and fluorescence approaches [93].Kinetic analysis of the terminator/antiterminator rearrangement and li-gand binding indicate that, in the absence of polymerase pausing, theassociation rate is sufficiently rapid to trap the terminator state at pro-posed intracellular preQ1 concentrations (30 μM). However, the authorsclearly state that without in vivo analysis, the kinetic versus thermody-namic issue in the context of transcription in the cell remains an openquestion [93].

7. Folding of the riboswitch

Since many riboswitches act co-transcriptionally, the ability to foldrapidly and efficiently is critical for the riboswitch function. First, theRNA must acquire the secondary and tertiary structure that definesthe global architecture of the aptamer domain. Second, the bindingpocket is formed during this phase and becomes competent for ligandbinding. Generally, in the apo-state, riboswitch binding pockets areweakly organized and exhibit a conformational ensemble, of whichonly a subset is productive for ligand binding, which in turn strongly in-fluences the association rate [51,66,94]. Furthermore, since the purinenucleobase is almost completely solvent inaccessible when bound tothe aptamer, there must be a conformational change in the three-wayjunction coupled to binding. Finally, since the P1 helixmust at least par-tially form to create the high-affinity aptamer for the ligand, if theriboswitch aptamer remains unbound then an alternative secondarystructure must form at its expense. Together, these events illustratethat the purine riboswitch is highly dynamic during the course of itsfunction and to fully understand how this RNA transduces intracellulareffector concentration into a regulatory response, these processes mustbe considered.

The folding of the aptamer domain has been investigated using anumber of experimental techniques including NMR [57,95,96], smFRET[52,56], chemical probing [51], fast fluorescence spectroscopy [97,98],single molecule force extension spectroscopy [91,99,100], as well asmolecular dynamics simulation [101–104]. While all of these tech-niques observe different aspects of the folding process, together theyyield a reasonably consistent model of the folding process. The best


models for acquisition of secondary structure come from singlemolecule force extension spectroscopy of the pbuE and add adenineriboswitch aptamers [91,100]. These studies reveal sequential forma-tion of the P2 and P3 hairpins, then the L2–L3 interaction, and finallyP1. Since only a few base pairs of P1 are required for moderate-affinityligand binding [47], it is likely that productive ligand binding starts tooccur prior to full formation of the P1 helix. smFRET studies indicatethat the L2–L3 interaction is dynamic and forms to a variable extent inthe apo-state in different purine riboswitch variants (see Section 3)[52,56]. It should be noted that these experiments investigate the fold-ing of the full length aptamer domain and that the observed foldingpathway may be different from a co-transcriptional folding pathway(see Section 7).

One of themost poorly understood aspects of bound state formationis the process of initial ligand docking with the three-way junction andthe coupled folding of RNA around the ligand. There is substantial dis-agreement about the degree of preorganization of the junction andhow the aptamer initially recognizes the appropriate ligand. In compet-ing NMR studies of the xpt riboswitch, one group finds that in theabsence of ligand the three-way junction lacks any indication of hydro-gen bonding that defines the bound state [57], while another arguesthat the core is moderately preorganized around the junction adjacentto P2/P3 [96]. It should be pointed out that in the latter study, mutationswere introduced into the P2 tune box artificially stabilizing this part ofthe RNA, which may have significantly influenced their observations.A more illuminating study by the Schwalbe group used time-resolveNMR methods to investigate ligand induced folding of the junctionusing photo caged hypoxanthine to synchronize binding [95]. Twosets of folding rates were observed: a faster group (t1/2 = 19–24 s) inthe junction and a slower set (t1/2 = 27–30 s) in the L2–L3 interaction.Along with other experiments indicating line broadening for non-specific ligands in the core, these data suggest a model in which the li-gand forms a low-affinity encounter complex with the core, followedby organization of the core around the specific ligand then followedby full stabilization of the L2–L3 interaction that locks P2 and P3 intoplace.

Using ultrafast multidimensional NMR methods, the Varani groupwas able to generate an even higher resolution model of adenine/magnesium induced folding of the add aptamer [105]. Theirmodel, gen-erally consistent with those based upon single molecule studies, clearlyreveals fast initial docking of adenine with the core (~16 s), while theL2–L3 interaction and a region of the P1 helix proximal to the ligandbinding site remain unfolded. Kinetically slower folding events includefull stabilization of the L2–L3 interaction (~28–~58 s) while P1 remainsflexible. Finally, the native structure is fully acquired after 2–3 min. Inparticular, these data highlight the direct role of the ligand in stabilizingthe P1 helix, a central feature of the alternative structural switch ofmany riboswitches.

The most elusive aspect of the folding process is the nature of the ini-tial docking event between the ligand and the three-way junction. Basedupon other kinetic and temperature-dependent chemical protection data,the Batey group proposed amodel inwhich the ligand initially dockswithnucleotide 74 in J3/1 and J2/3 folds around the ligand to form the finalcomplex [51]. Independent molecular dynamics (MD) simulations are indisagreement on this point, either proposing models in which the ligandinitially docks with nucleotide 51 [102,103], or that 74 is the initialdocking site [104,106]. However, all of these studies must be taken withsome caution as the force fields used for these calculations are still awork in progress and struggle to accurately describe experimental obser-vations of even simple RNAmotifs [107–109]. Nonetheless, these simula-tions present interesting hypotheses that may become the basis forfurther experiments that will address this issue regarding both the initialligand recognition and specificity for the cognate effector.

To fully understand the folding pathway of the purine riboswitchand its relationship to regulatory activity, optimal folding needs to beobserved in the context of transcription. Currently this is not technically




possible using ensemble methods because it would require that theentire population be nearly synchronous and stochastic events such asthe entry and exit from programmed pauses frustrate this approach.To address this issue, Frieda and Block used force extension spectrosco-py to unfold and refold nascent RNA pbuE riboswitch transcripts in thepresence of varying concentrations of adenine [99]. In this experiment,Escherichia coli RNAP stalled downstream of the promoter was attachedto a bead in one optical trap, while the 5′-end of the RNA transcript washybridized to a DNA handle attached to a second trapped bead.Transcription was reinitiated and allowed to proceed to the end of thetranscription template where further elongation was blocked bystreptavidin bound to the biotinylated DNA (a good review of thetools and experimental methods of force extension spectroscopy as ap-plied to RNA is given by [110]). If transcriptional termination occurs atthe Rho-independent terminator, the RNAP disengages and the com-plex falls apart such that the two beads are no longer tethered. Sincethese experiments are capable of observing transcriptional events inreal time, they yield a complete co-transcriptional folding landscape ofan RNA—the first to be achieved (Fig. 5). Like previous studies of theaptamer alone, the earliest event is formation of the P2 helix. Shortlyafter the polymerase cleared the aptamer domain sequence, theaptamer domain obtained its global architecture. If adenine binds tothe aptamer, the ligand stabilizes this structure against invasion fromthe terminator sequence long enough to enable RNAP to escape pastthe polyuridine tract of the intrinsic terminator. If adenine does notbind, then the terminator stem–loop forms rapidly at the expense ofthe P1 helix once its sequence had been fully transcribed. Further,these experiments clearly demonstrate kinetic control of the pbuEriboswitch. In the vast majority of observations, the aptamer folds andbinds adenine only once before the polymerase escapes past the termi-nator. Therefore, the aptamer does not have sufficient time to fullyequilibrate (requiring multiple samplings of the bound/unbound stateson average) during the transcription process. Finally, in these experi-ments the polymerase was not observed to pause as expected at thetwo poly-uridine tracts. The absence of polymerase stalling was

Fig. 5. Single nucleotide resolution co-transcriptional foldingmap of thepurine riboswitch. The sextension spectroscopywith secondary structural elements in the aptamer highlighted as shadscription trajectory as indicated by the cartoons. For individual experiments that read-throughwthis riboswitch is kinetically driven.Figure adapted from [92].


interpreted as differences in one of several experimental parametersand merits further investigation since the Lafontaine group uncovereda clear influence on pausing in the pbuE riboswitch [86].

8. Why riboswitches?

Given the pervasiveness of riboswitches across bacteria, the diversi-ty of small molecules recognized, and their regulation of essentialmetabolite biosynthesis [111], it is clear that these RNAs confer a signif-icant advantage in the maintenance of normal cellular homeostasis.With their numbers outpacing that of known protein repressors insome Firmicutes [1], these RNAs must possess some advantage overtheir protein counterparts. One reason might be that the RNA has anability to integrate multiple sensory inputs into a single, appropriateregulatory response. Put another way, the regulatory response is acomplex function of not only the effector concentration, but also otherintracellular conditions. As discussed above, the intracellular NTP con-centration affects the transcription rate, which for kinetically controlledriboswitches alters their T50. A second factor which can significantlyfluctuate in some bacteria is magnesium ion concentration, though itis typically tightly regulated [112]. Another class of RNA regulatory de-vices called “thermosensors” exploits the thermal stability of RNA sec-ondary structure [113–115]. Even more recently, an RNA pH sensorwas characterized that represses expression of the alx locus at elevatedextracellular pH [116]. Thus, the regulatory landscape of a specificriboswitch can be very complex allowing the RNA to respond in realtime to a variety of intra- and extracellular cues [90].

An example of how RNAmodulates regulation in response to an en-vironmental factorwas recently described by the Schwalbe group [117].They correctly note that a strictly two-state RNA would have difficultymaintaining a strong regulatory response under varying temperatureconditions. At low temperatures, the RNA's affinity for the ligand isgreater than at high temperatures such that a regulatory RNA wouldonly be efficient at low temperature. However, binding is not a twostate process, as has been shown in a number of studies. An earlier

equenceof theB. subtilis pbuE adenine responsive riboswitchused for singlemolecule forceed boxes. Secondary structure acquisition and ligand binding was observed along the tran-as observed, binding of adeninewas observed to occur only once, supporting the idea that




study of the xpt–pbuX guanine riboswitch aptamer showed that over abroad temperature range (5–60 °C) the free energy of binding (ΔG°)is fairly constant as measured by isothermal titration calorimetry [43].But these same experiments revealed a marked temperature depen-dence upon the heat capacity (∂ΔH°/∂T), which is generally interpretedas temperature-dependent changes in the ensemble of conformers ofthe apo-state. Thus, the population of RNA competent to bind ligandchanges as a function of temperature, which in turn is likely to influenceits regulatory properties. Similar observations were made for an SAH-binding riboswitch [118]. As many other riboswitch aptamers experi-ence multiple apo-state conformations, this is likely a general feature[94].

In the study by the Schwalbe group of a nearly full length addriboswitch (a small truncation of seven nucleotides was used at the5′-end), the structure of two distinct apo-form conformers was eluci-dated by NMR. The first is a native-like state that is competent to bindadenine, while the second has an alternative secondary structure thatdestroys the P1 helix. Importantly, these two states are bistable suchthat they interconvert in response to temperature changes. Using acombination of binding analysis and an in vitro transcription/translationassay, they demonstrate that the equilibrium population of the apo-forms changes between 10 and 37 °C in such a way that enables theregulatory response to remain robust. This has important consequencesfor V. vulnificus as it experiences both temperatures depending uponwhether it is living freely in a marine environment (10 °C) or infectinga human host (37 °C), similar to a previously characterized RNAthermosensor in Listeria monocytogenes [114]. Given the diversity ofenvironments and how rapidly conditions can change, it is not surpris-ing that RNA-mediated regulatory mechanisms which are nativelypredisposed to be exquisitely sensitive to these factors are extensively uti-lized to regulate gene expression. Strikingly, examples of temperature-dependent riboregulation of virulence factors were recently describedin Yersinia species [119] and Neisseria meningitidis [120].

9. Application of purine riboswitches

RNA-based sensors are increasingly seen as important tools for syn-thetic biology, of which riboswitches represent Nature's solution to theimplementation of such devices [121–124]. One important applicationwhere riboswitches can make a significant contribution is the develop-ment of new regulatory elements for inducible gene expression. Whileprotein-based regulation is typically used for this task, a number of spe-cific needs have arisen in synthetic biology that demand new regulatoryelements for which RNA is particularly well suited. These include differ-ential induction ofmultiple genes in a pathway or inmulti-protein com-plexes and minimizing cross-talk between regulatory elements.

To address these issues, theMicklefield group has developed orthog-onal riboswitches responding to non-natural small molecules using theadd adenine-responsive riboswitch as a scaffold [125]. To find variantsof the add riboswitch responsive to alternative compounds, all possiblenucleotide combinations of positions 47 and 51 (Fig. 1C) were clonedupstream of the CAT gene and each was screened against a chemical li-brary of ~80 heterocycles for conferring chloramphenicol resistance.One variant (U47C, U51C) was found to specifically bind ammeline(4,6-diamino-2-hydroxy-1,3,5-triazine) and resulted in a regulatoryelement capable of robustly inducing varying levels of protein expres-sion in response to its effector. This pioneering study suggests thatreengineering of naturally occurring riboswitch aptamer domains is aviable solution to the creation of genetic control elements that functionin a variety of contexts.

To demonstrate how orthogonal riboswitches can assist in the regu-latory control ofmultiple genes, the authors created dual and single pro-moter operons in which the natural add and the M6 translational “ON”riboswitches regulate the DSRed and eGFP genes, respectively [126]. Inboth examples, each gene was specifically induced by its cognate effec-tor (2-aminopurine for add and ammeline for M6). For the dual


promoter system, therewas no cross-talk between the riboswitches de-spite their sequence differing by only two point mutations in the ligandbinding core. In contrast, under a single promoter in which the M6riboswitch is in the leader sequence and add is an intercistronic controlelement of the DsRed:eGFP cassette, add exhibits a moderate depen-dence on both riboswitch effector ligands for maximal expression. Thedownside to this approach is that the fold induction for theseriboswitches (10–25 fold) is lower than some protein systems.

Synthetic riboswitches have also been created throughmodificationof the expression platform. It is well appreciated that aptamer domainsof riboswitches, like their SELEX-derived counterparts, are highlycomposable such that they can function in a variety of contexts includ-ing synthetic riboswitches, aptazymes (aptamer inducible ribozymes),and fluorescent biosensors [123]. However, modularity of expressionplatforms had not been demonstrated until recently [127,128]. Insome riboswitches, the sequence required for high affinity ligand bind-ing to the aptamer domain does not overlap with the alternative sec-ondary structural switch, particularly those in the P1 helix consideredto be central to interdomain communication. An example of such an ar-rangement is the B. subtilis metE transcriptional “OFF” riboswitch, inwhich the two base pairs essential for S-adenosylmethionine (SAM)binding to the aptamer are localized to the 3′-side of the P1 helixwhile nucleotides that participate in the P1/P-AT switch are localizedto the 5′-side (Fig. 6A) [127]. By defining the boundary of the two do-mains as the region of P1 where these elements meet, the aptamercan be replaced by a number of alternative sensors, both natural andin vitro selected with the resultant chimeras functional both in vitro(Fig. 6B) and in vivo.

From a mechanistic perspective, the adaptability of select ex-pression platforms suggests that these RNAs exploit an “encodedco-transcriptional” folding process [129]. In this model of RNA fold-ing, the secondary structure adopted by the RNA is a function of therelative thermodynamic stability of competing helical elements as wellas their 5′-to-3′ polarity. This principle was first demonstrated in fold-ing of small model RNAs that have isoenergetic, mutually exclusivestructures [129]. For these RNAs, the folding landscape is directed bythe sequential ordering of the helical elements. In transcriptional“OFF” expression platforms ordering of the P1/P-AT elements remainfixed, but the relative stabilities of the helical elements are variable[130]. In the absence of ligand, P-AT is the more stable element, en-abling its formation to dominate over P1. Conversely, ligand to a sitein the aptamer domain that is typically adjacent to the 3′-side of theP1, increases the relative stability of P1 over P-AT, leading to formationof the downstream intrinsic terminator. Thus, the expression platformdoes not necessarily care about the particulars of the ligand–aptamercomplex—only that it alters the stability of P1, which is why many dis-tinct aptamer domains can efficiently direct ligand-dependent tran-scriptional termination of the metE expression platform [127]. Moreimportantly, since SELEX-derived aptamers can be incorporated intofunctional chimeric riboswitches, there is nothing special about biolog-ical aptamers that enable them to interfacewith anRNA switch. Instead,inmost cases, the aptamermerely has to position the ligand binding site(or a coupled conformational change in the RNA) to influence P1stability.

Amore useful modular switch is one that can induce gene expressionin response to ligand binding, like theM6 riboswitch [125,126]. However,most of the “ON” switches surveyed for potential modularity have signif-icant sequence overlap between the two domains, preventing their sim-ple decoupling at a defined boundary like the metE riboswitch [130].Instead, the well-characterized adenine-responsive pbuE riboswitchfrom B. subtilis was re-engineered to enable its coupling to differentaptamers [130]. In the natural pbuE riboswitch, the terminator (P–T) in-vades into L3, such that the alternative secondary structural switch in-cludes sequence elements of the aptamer necessary for both ligandbinding and tertiary structure formation. To decouple the two domains,a short sequence was added to the 3′-side of the P1 helix that prevented



Fig. 6. Synthesis of chimeric riboswitches from biological parts. (A) In order for the expression platform to be used as a modular part, all of the sequences required for the alternative sec-ondary structural switchmust be distinct from the sequence that defines theminimal aptamer domain (yellow). Inmany riboswitches, the key element is the “switching sequence” (red),the region of the expression platform (cyan) sharedbetween the two alternative structures. (B) Anexample of a functional chimeric riboswitch. The top shows thewild type B. subtilis metESAM-responsive riboswitch undergoing transcriptional attenuation as a function of increasing ligand concentration in a single-turnover transcription assay (RT, read through product; T,terminated product). The middle shows that a chimera of the B. subtilis xpt–pbuX guanine binding aptamer domain and the B. subtilis metE expression platform is no longer SAM-responsive, but is strongly responsive to guanine (bottom).Figure adapted from [118].


the invasion of P–T into the aptamer, mimicking the arrangement ob-served inmetE. This engineered variant of the pbuE expression platformis able to host a variety of aptamers to turn on gene expression. Thesestudies have only explored a small number of expression platforms,and it may be that even more useful regulatory modules can be createdfrom translational attenuators or even eukaryotic riboswitches that regu-late splicing or mRNA stability.

10. Conclusion

The purine riboswitch, and particularly its aptamer domain, has be-come an invaluable model system for exploring facets of RNA folding,small molecule recognition (reviewed more extensively in [40]), andstructure. From this extensive body of work, we are beginning to eluci-date the foundational principles of how these RNAs function to efficient-ly regulate gene expression across a highly diverse set of organisms thatexperience a broad spectrum of environmental conditions. The door isnow opening to engineering naturally based sensors that are modular,adaptable, and flexible in their application with the benefits stemmingfrom the vast amount of knowledge gained by studies of the purineriboswitch. Nonetheless, there are still many aspects of these RNAsthat remain to be elucidated, in particular a level of understanding ofthe expression platform that rivals the aptamer domain as well as abetter appreciation of how these RNAs function in the cellular context.

Acknowledgements

Funding for this work was provided by a grant from the NationalInstitutes of Health to R.T.B. (GM 073850).

References

[1] R.R. Breaker, Riboswitches and the RNA world, Cold Spring Harb. Perspect. Biol. 4(2012).


[2] C. Lu, A.M. Smith, R.T. Fuch, F. Ding, K. Rajashankar, T.M. Henkin, A. Ke, Crystalstructure of the SAM-III/S(MK) riboswitch reveal the SAM-dependent translationinhibition mechanism, Nat. Struct. Mol. Biol. 15 (2008) 1076–1083.

[3] J.E. Johnson Jr., F.E. Reyes, J.T. Polaski, R.T. Batey, B12 cofactors directly stabilize anmRNA regulatory switch, Nature 492 (2012) 133–137.

[4] R.K. Montange, R.T. Batey, Structure of the S-adenosylmethionine riboswitch regu-latory mRNA element, Nature 441 (2006) 1172–1175.

[5] S. Thore, M. Leibundgut, N. Ban, Structure of the eukaryotic thiamine pyrophos-phate riboswitch with its regulatory ligand, Science 312 (2006) 1208–1211.

[6] A. Serganov, A. Polonskaia, A.T. Phan, R.R. Breaker, D.J. Patel, Structural basis forgene regulation by a thiamine pyrophosphate-sensing riboswitch, Nature 441(2006) 1167–1171.

[7] D.J. Klein, T.E. Edwards, A.R. Ferre-D'Amare, Cocrystal structure of a class I preQ1riboswitch reveals a pseudoknot recognizing an essential hypermodifiednucleobase, Nat. Struct. Mol. Biol. 16 (2009) 343–344.

[8] M. Kang, R. Peterson, J. Feigon, Structural insights into riboswitch control of thebiosynthesis of queuosine, a modified nucleotide found in the anticodon of tRNA,Mol. Cell 33 (2009) 784–790.

[9] R.C. Spitale, A.T. Torelli, J. Krucinska, V. Bandarian, J.E. Wedekind, The structuralbasis for recognition of the PreQ0 metabolite by an unusually small riboswitchaptamer domain, J. Biol. Chem. 284 (2009) 11012–11016.

[10] R.T. Batey, S.D. Gilbert, R.K. Montange, Structure of a natural guanine-responsiveriboswitch complexed with the metabolite hypoxanthine, Nature 432 (2004)411–415.

[11] A. Serganov, Y.R. Yuan, O. Pikovskaya, A. Polonskaia, L. Malinina, A.T. Phan, C.Hobartner, R. Micura, R.R. Breaker, D.J. Patel, Structural basis for discriminative reg-ulation of gene expression by adenine- and guanine-sensing mRNAs, Chem. Biol.11 (2004) 1729–1741.

[12] L.C. Christiansen, S. Schou, P. Nygaard, H.H. Saxild, Xanthine metabolism in Bacillussubtilis: characterization of the xpt–pbuX operon and evidence for purine- andnitrogen-controlled expression of genes involved in xanthine salvage and catabo-lism, J. Bacteriol. 179 (1997) 2540–2550.

[13] R. Landick, C. Yanofsky, Stability of an RNA secondary structure affects in vitro tran-scription pausing in the trp operon leader region, J. Biol. Chem. 259 (1984)1550–1555.

[14] F.J. Grundy, T.M. Henkin, The S box regulon: a new global transcription terminationcontrol system for methionine and cysteine biosynthesis genes in Gram-positivebacteria, Mol. Microbiol. 30 (1998) 737–749.

[15] J. Miranda-Rios, M. Navarro, M. Soberon, A conserved RNA structure (thi box) is in-volved in regulation of thiamin biosynthetic gene expression in bacteria, Proc. Natl.Acad. Sci. U. S. A. 98 (2001) 9736–9741.

[16] M.S. Gelfand, A.A. Mironov, J. Jomantas, Y.I. Kozlov, D.A. Perumov, A conserved RNAstructure element involved in the regulation of bacterial riboflavin synthesis genes,Trends Genet. 15 (1999) 439–442.

[17] X. Nou, R.J. Kadner, Adenosylcobalamin inhibits ribosome binding to btuB RNA,Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 7190–7195.


http://refhub.elsevier.com/S1874-9399(14)00035-2/rf0650


















































[18] G.D. Stormo, Y. Ji, Do mRNAs act as direct sensors of small molecules to controltheir expression? Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 9465–9467.

[19] V. Epshtein, A.S. Mironov, E. Nudler, The riboswitch-mediated control of sulfurmetabolism in bacteria, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 5052–5056.

[20] B.A. McDaniel, F.J. Grundy, I. Artsimovitch, T.M. Henkin, Transcription terminationcontrol of the S box system: direct measurement of S-adenosylmethionine by theleader RNA, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 3083–3088.

[21] A.S. Mironov, I. Gusarov, R. Rafikov, L.E. Lopez, K. Shatalin, R.A. Kreneva, D.A.Perumov, E. Nudler, Sensing small molecules by nascent RNA: a mechanism tocontrol transcription in bacteria, Cell 111 (2002) 747–756.

[22] A. Nahvi, N. Sudarsan, M.S. Ebert, X. Zou, K.L. Brown, R.R. Breaker, Genetic controlby a metabolite binding mRNA, Chem. Biol. 9 (2002) 1043.

[23] W.C. Winkler, S. Cohen-Chalamish, R.R. Breaker, An mRNA structure that controlsgene expression by binding FMN, Proc. Natl. Acad. Sci. U. S. A. 99 (2002)15908–15913.

[24] M. Mandal, B. Boese, J.E. Barrick, W.C. Winkler, R.R. Breaker, Riboswitches controlfundamental biochemical pathways in Bacillus subtilis and other bacteria, Cell113 (2003) 577–586.

[25] M. Mandal, R.R. Breaker, Gene regulation by riboswitches, Nat. Rev. Mol. Cell Biol. 5(2004) 451–463.

[26] J.N. Kim, A. Roth, R.R. Breaker, Guanine riboswitch variants from Mesoplasmaflorum selectively recognize 2′-deoxyguanosine, Proc. Natl. Acad. Sci. U. S. A. 104(2007) 16092–16097.

[27] M. de la Pena, D. Dufour, J. Gallego, Three-way RNA junctions with remote tertiarycontacts: a recurrent and highly versatile fold, RNA 15 (2009) 1949–1964.

[28] A. Lescoute, E. Westhof, Topology of three-way junctions in folded RNAs, RNA 12(2006) 83–93.

[29] V. Delfosse, P. Bouchard, E. Bonneau, P. Dagenais, J.F. Lemay, D.A. Lafontaine, P.Legault, Riboswitch structure: an internal residue mimicking the purine ligand,Nucleic Acids Res. 38 (2010) 2057–2068.

[30] O. Pikovskaya, A. Polonskaia, D.J. Patel, A. Serganov, Structural principles of nucleosideselectivity in a 2′-deoxyguanosine riboswitch, Nat. Chem. Biol. 7 (2011) 748–755.

[31] M. Mandal, R.R. Breaker, Adenine riboswitches and gene activation by disruption ofa transcription terminator, Nat. Struct. Mol. Biol. 11 (2004) 29–35.

[32] S.W. Burge, J. Daub, R. Eberhardt, J. Tate, L. Barquist, E.P. Nawrocki, S.R. Eddy, P.P.Gardner, A. Bateman, Rfam 11.0: 10 years of RNA families, Nucleic Acids Res. 41(2013) D226–D232.

[33] S. Griffiths-Jones, A. Bateman, M. Marshall, A. Khanna, S.R. Eddy, Rfam: an RNAfamily database, Nucleic Acids Res. 31 (2003) 439–441.

[34] S.D. Gilbert, F.E. Reyes, A.L. Edwards, R.T. Batey, Adaptive ligand binding by the pu-rine riboswitch in the recognition of guanine and adenine analogs, Structure 17(2009) 857–868.

[35] A.L. Edwards, R.T. Batey, A structural basis for the recognition of 2′-deoxyguanosine by the purine riboswitch, J. Mol. Biol. 385 (2009) 938–948.

[36] J.J. Trausch, R.T. Batey, A disconnect between high-affinity binding and efficientregulation by antifolates and purines in the tetrahydrofolate riboswitch, A discon-nect between high-affinity binding and efficient regulation by antifolates andpurines in the tetrahydrofolate riboswitch, Chem. Biol. 21 (2014) 205–216.

[37] J.A. Liberman, M. Salim, J. Krucinska, J.E. Wedekind, Structure of a class II preQ1riboswitch reveals ligand recognition by a new fold, Nat. Chem. Biol. 9 (2013)353–355.

[38] G.R. Zimmermann, R.D. Jenison, C.L. Wick, J.P. Simorre, A. Pardi, Interlocking struc-tural motifs mediate molecular discrimination by a theophylline-binding RNA, Nat.Struct. Biol. 4 (1997) 644–649.

[39] R.T. Batey, Recognition of S-adenosylmethionine by riboswitches, WileyInterdiscip. Rev. RNA 2 (2011) 299–311.

[40] R.T. Batey, Structure and mechanism of purine-binding riboswitches, Q. Rev.Biophys. 45 (2012) 345–381.

[41] Q. Vicens, T.R. Cech, Atomic level architecture of group I introns revealed, TrendsBiochem. Sci. 31 (2006) 41–51.

[42] P. Daldrop, F.E. Reyes, D.A. Robinson, C.M. Hammond, D.M. Lilley, R.T. Batey, R.Brenk, Novel ligands for a purine riboswitch discovered by RNA–ligand docking,Chem. Biol. 18 (2011) 324–335.

[43] S.D. Gilbert, C.D. Stoddard, S.J.Wise, R.T. Batey, Thermodynamic and kinetic charac-terization of ligand binding to the purine riboswitch aptamer domain, J. Mol. Biol.359 (2006) 754–768.

[44] J.N. Kim, K.F. Blount, I. Puskarz, J. Lim, K.H. Link, R.R. Breaker, Design and antimicro-bial action of purine analogues that bind guanine riboswitches, ACS Chem. Biol. 4(2009) 915–927.

[45] S.D. Gilbert, S.J. Mediatore, R.T. Batey, Modified pyrimidines specifically bind thepurine riboswitch, J. Am. Chem. Soc. 128 (2006) 14214–14215.

[46] J. Mulhbacher, E. Brouillette, M. Allard, L.C. Fortier, F. Malouin, D.A. Lafontaine,Novel riboswitch ligand analogs as selective inhibitors of guanine-relatedmetabol-ic pathways, PLoS Pathog. 6 (2010) e1000865.

[47] J.F. Lemay, D.A. Lafontaine, Core requirements of the adenine riboswitch aptamerfor ligand binding, RNA 13 (2007) 339–350.

[48] C.D. Stoddard, J. Widmann, J.J. Trausch, J.G. Marcano-Velazquez, R. Knight,R.T. Batey, Nucleotides adjacent to the ligand-binding pocket are linkedto activity tuning in the purine riboswitch, J. Mol. Biol. 425 (2013)1596–1611.

[49] J.K.Wickiser, M.T. Cheah, R.R. Breaker, D.M. Crothers, The kinetics of ligand bindingby an adenine-sensing riboswitch, Biochemistry 44 (2005) 13404–13414.

[50] K.F. Blount, R.R. Breaker, Riboswitches as antibacterial drug targets, Nat.Biotechnol. 24 (2006) 1558–1564.

[51] C.D. Stoddard, S.D. Gilbert, R.T. Batey, Ligand-dependent folding of the three-wayjunction in the purine riboswitch, RNA 14 (2008) 675–684.


[52] J.F. Lemay, J.C. Penedo, R. Tremblay, D.M. Lilley, D.A. Lafontaine, Folding of theadenine riboswitch, Chem. Biol. 13 (2006) 857–868.

[53] S.D. Gilbert, C.E. Love, A.L. Edwards, R.T. Batey, Mutational analysis of the purineriboswitch aptamer domain, Biochemistry 46 (2007) 13297–13309.

[54] P. Nissen, J.A. Ippolito, N. Ban, P.B. Moore, T.A. Steitz, RNA tertiary interactions inthe large ribosomal subunit: the A-minor motif, Proc. Natl. Acad. Sci. U. S. A. 98(2001) 4899–4903.

[55] E.A. Doherty, R.T. Batey, B. Masquida, J.A. Doudna, A universal mode of helix pack-ing in RNA, Nat. Struct. Biol. 8 (2001) 339–343.

[56] M.D. Brenner, M.S. Scanlan, M.K. Nahas, T. Ha, S.K. Silverman, Multivector fluores-cence analysis of the xpt guanine riboswitch aptamer domain and the conforma-tional role of guanine, Biochemistry 49 (2010) 1596–1605.

[57] J. Noeske, J. Buck, B. Furtig, H.R. Nasiri, H. Schwalbe, J. Wohnert, Interplay of ‘in-duced fit’ and preorganization in the ligand induced folding of the aptamer domainof the guanine binding riboswitch, Nucleic Acids Res. 35 (2007) 572–583.

[58] S.D. Gilbert, R.T. Batey, Riboswitches: fold and function, Chem. Biol. 13 (2006)805–807.

[59] J. Noeske, C. Richter, M.A. Grundl, H.R. Nasiri, H. Schwalbe, J. Wohnert, An intermo-lecular base triple as the basis of ligand specificity and affinity in the guanine- andadenine-sensing riboswitch RNAs, Proc. Natl. Acad. Sci. U. S. A. 102 (2005)1372–1377.

[60] J. Noeske, H. Schwalbe, J. Wohnert, Metal–ion binding and metal–ion induced fold-ing of the adenine-sensing riboswitch aptamer domain, Nucleic Acids Res. 35(2007) 5262–5273.

[61] J.F. Lemay, D.A. Lafontaine, The adenine riboswitch: a new gene regulation mech-anism, Med. Sci. (Paris) 22 (2006) 1053–1059.

[62] J.C. Bowman, T.K. Lenz, N.V. Hud, L.D. Williams, Cations in charge: magnesium ionsin RNA folding and catalysis, Curr. Opin. Struct. Biol. 22 (2012) 262–272.

[63] D.E. Draper, A guide to ions and RNA structure, RNA 10 (2004) 335–343.[64] T.E. Edwards, A.R. Ferre-D'Amare, Crystal structures of the thi-box riboswitch

bound to thiamine pyrophosphate analogs reveal adaptive RNA-small moleculerecognition, Structure 14 (2006) 1459–1468.

[65] A. Serganov, L. Huang, D.J. Patel, Coenzyme recognition and gene regulation by aflavin mononucleotide riboswitch, Nature 458 (2009) 233–237.

[66] Q. Vicens, E. Mondragon, R.T. Batey, Molecular sensing by the aptamer domain ofthe FMN riboswitch: a general model for ligand binding by conformational selec-tion, Nucleic Acids Res. 39 (2011) 8586–8598.

[67] A. Ren, K.R. Rajashankar, D.J. Patel, Fluoride ion encapsulation by Mg2+ ions andphosphates in a fluoride riboswitch, Nature 486 (2012) 85–89.

[68] E.B. Butler, Y. Xiong, J. Wang, S.A. Strobel, Structural basis of cooperative ligandbinding by the glycine riboswitch, Chem. Biol. 18 (2011) 293–298.

[69] C.E. Dann III, C.A.Wakeman, C.L. Sieling, S.C. Baker, I. Irnov,W.C. Winkler, Structureand mechanism of a metal-sensing regulatory RNA, Cell 130 (2007) 878–892.

[70] L. Huang, A. Serganov, D.J. Patel, Structural insights into ligand recognition by asensing domain of the cooperative glycine riboswitch, Mol. Cell 40 (2010)774–786.

[71] R.T. Batey, J.A. Doudna, Structural and energetic analysis of metal ions essential toSRP signal recognition domain assembly, Biochemistry 41 (2002) 11703–11710.

[72] J. Buck, J. Noeske, J. Wohnert, H. Schwalbe, Dissecting the influence of Mg2+ on 3Darchitecture and ligand-binding of the guanine-sensing riboswitch aptamerdomain, Nucleic Acids Res. 38 (2010) 4143–4153.

[73] D. Leipply, D.E. Draper, Effects of Mg2+ on the free energy landscape for folding apurine riboswitch RNA, Biochemistry 50 (2011) 2790–2799.

[74] D. Lambert, D. Leipply, R. Shiman, D.E. Draper, The influence of monovalent cationsize on the stability of RNA tertiary structures, J. Mol. Biol. 390 (2009) 791–804.

[75] R. Shiman, D.E. Draper, Stabilization of RNA tertiary structure by monovalentcations, J. Mol. Biol. 302 (2000) 79–91.

[76] D.E. Draper, V.K. Misra, RNA shows its metal, Nat. Struct. Biol. 5 (1998) 927–930.[77] S. Basu, R.P. Rambo, J. Strauss-Soukup, J.H. Cate, A.R. Ferre-D'Amare, S.A. Strobel, J.A.

Doudna, A specific monovalent metal ion integral to the AA platform of the RNAtetraloop receptor, Nat. Struct. Biol. 5 (1998) 986–992.

[78] J.E. Barrick, K.A. Corbino, W.C. Winkler, A. Nahvi, M. Mandal, J. Collins, M. Lee, A.Roth, N. Sudarsan, I. Jona, J.K. Wickiser, R.R. Breaker, New RNA motifs suggest anexpanded scope for riboswitches in bacterial genetic control, Proc. Natl. Acad. Sci.U. S. A. 101 (2004) 6421–6426.

[79] S. Blouin, R. Chinnappan, D.A. Lafontaine, Folding of the lysine riboswitch: impor-tance of peripheral elements for transcriptional regulation, Nucleic Acids Res. 39(2011) 3373–3387.

[80] M. Mandal, M. Lee, J.E. Barrick, Z. Weinberg, G.M. Emilsson, W.L. Ruzzo, R.R. Break-er, A glycine-dependent riboswitch that uses cooperative binding to control geneexpression, Science 306 (2004) 275–279.

[81] E.M. Sherman, J. Esquiaqui, G. Elsayed, J.D. Ye, An energetically beneficial leader–linker interaction abolishes ligand-binding cooperativity in glycine riboswitches,RNA 18 (2012) 496–507.

[82] W. Kladwang, F.C. Chou, R. Das, Automated RNA structure prediction uncovers akink-turn linker in double glycine riboswitches, J. Am. Chem. Soc. 134 (2012)1404–1407.

[83] N.J. Baird, A.R. Ferre-D'Amare, Modulation of quaternary structure and enhance-ment of ligand binding by the K-turn of tandem glycine riboswitches, RNA 19(2013) 167–176.

[84] J.R. Mellin, T. Tiensuu, C. Becavin, E. Gouin, J. Johansson, P. Cossart, A riboswitch-regulated antisense RNA in Listeria monocytogenes, Proc. Natl. Acad. Sci. U. S. A.110 (2013) 13132–13137.

[85] K. Hollands, S. Proshkin, S. Sklyarova, V. Epshtein, A. Mironov, E. Nudler, E.A.Groisman, Riboswitch control of Rho-dependent transcription termination, Proc.Natl. Acad. Sci. U. S. A. 109 (2012) 5376–5381.






















































































































































































[86] J.F. Lemay, G. Desnoyers, S. Blouin, B. Heppell, L. Bastet, P. St-Pierre, E. Masse, D.A.Lafontaine, Comparative study between transcriptionally- and translationally-acting adenine riboswitches reveals key differences in riboswitch regulatorymech-anisms, PLoS Genet. 7 (2011) e1001278.

[87] R. Rieder, K. Lang, D. Graber, R. Micura, Ligand-induced folding of the adenosinedeaminase A-riboswitch and implications on riboswitch translational control,Chembiochem 8 (2007) 896–902.

[88] J.K. Wickiser, W.C. Winkler, R.R. Breaker, D.M. Crothers, The speed of RNA tran-scription and metabolite binding kinetics operate an FMN riboswitch, Mol. Cell18 (2005) 49–60.

[89] J.C. McDowell, J.W. Roberts, D.J. Jin, C. Gross, Determination of intrinsic transcrip-tion termination efficiency by RNA polymerase elongation rate, Science 266(1994) 822–825.

[90] A.D. Garst, E.B. Porter, R.T. Batey, Insights into the regulatory landscape of thelysine riboswitch, J. Mol. Biol. 423 (2012) 17–33.

[91] K. Neupane, H. Yu, D.A. Foster, F. Wang, M.T. Woodside, Single-molecule forcespectroscopy of the add adenine riboswitch relates folding to regulatory mecha-nism, Nucleic Acids Res. 39 (2011) 7677–7687.

[92] K. McGary, E. Nudler, RNA polymerase and the ribosome: the close relationship,Curr. Opin. Microbiol. 16 (2013) 112–117.

[93] U. Rieder, C. Kreutz, R. Micura, Folding of a transcriptionally acting preQ1riboswitch, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 10804–10809.

[94] J.E.Wedekind, The apo riboswitch as amolecular hydra, Structure 18 (2010) 757–758.[95] J. Buck, B. Furtig, J. Noeske, J. Wohnert, H. Schwalbe, Time-resolved NMR methods

resolving ligand-induced RNA folding at atomic resolution, Proc. Natl. Acad. Sci. U.S. A. 104 (2007) 15699–15704.

[96] O.M. Ottink, S.M. Rampersad, M. Tessari, G.J. Zaman, H.A. Heus, S.S. Wijmenga,Ligand-induced folding of the guanine-sensing riboswitch is controlled by acombined predetermined induced fit mechanism, RNA 13 (2007) 2202–2212.

[97] S. Eskandari, O. Prychyna, J. Leung, D. Avdic, M.A. O'Neill, Ligand-directed dynamicsof adenine riboswitch conformers, J. Am. Chem. Soc. 129 (2007) 11308–11309.

[98] O. Prychyna, M.S. Dahabieh, J. Chao, M.A. O'Neill, Sequence-dependent folding andunfolding of ligand-bound purine riboswitches, Biopolymers 91 (2009) 953–965.

[99] K.L. Frieda, S.M. Block, Direct observation of cotranscriptional folding in an adenineriboswitch, Science 338 (2012) 397–400.

[100] W.J. Greenleaf, K.L. Frieda, D.A. Foster, M.T. Woodside, S.M. Block, Direct observa-tion of hierarchical folding in single riboswitch aptamers, Science 319 (2008)630–633.

[101] P.H. Nguyen, P. Derreumaux, G. Stock, Energy flow and long-range correlations inguanine-binding riboswitch: a nonequilibrium molecular dynamics study, J. Phys.Chem. B 113 (2009) 9340–9347.

[102] A. Villa, J. Wohnert, G. Stock, Molecular dynamics simulation study of the bindingof purine bases to the aptamer domain of the guanine sensing riboswitch, NucleicAcids Res. 37 (2009) 4774–4786.

[103] M. Sharma, G. Bulusu, A. Mitra, MD simulations of ligand-bound and ligand-freeaptamer: molecular level insights into the binding and switching mechanism ofthe add A-riboswitch, RNA 15 (2009) 1673–1692.

[104] Z. Gong, Y. Zhao, C. Chen, Y. Xiao, Role of ligand binding in structural organizationof add A-riboswitch aptamer: a molecular dynamics simulation, J. Biomol. Struct.Dyn. 29 (2011) 403–416.

[105] M.K. Lee, M. Gal, L. Frydman, G. Varani, Real-time multidimensional NMR followsRNA folding with second resolution, Proc. Natl. Acad. Sci. U. S. A. 107 (2010)9192–9197.

[106] U.D. Priyakumar, A.D. MacKerell Jr., Role of the adenine ligand on the stabilizationof the secondary and tertiary interactions in the adenine riboswitch, J. Mol. Biol.396 (2010) 1422–1438.

[107] P. Banáš, D. Hollas, M. Zgarbová, P. Jurečka,M. Orozco, T.E. Cheatham III, J.i. Šponer,M.Otyepka, Performance of molecular mechanics force fields for RNA simulations:stability of UUCG and GNRA hairpins, J. Chem. Theory Comput. 6 (2010) 3836–3849.


[108] M.A. Ditzler, M. Otyepka, J. Sponer, N.G. Walter, Molecular dynamics and quantummechanics of RNA: conformational and chemical change we can believe in, Acc.Chem. Res. 43 (2010) 40–47.

[109] S.E.McDowell, N.a. Špačková, J. Šponer, N.G.Walter,Molecular dynamics simulationsof RNA: an in silico single molecule approach, Biopolymers 85 (2007) 169–184.

[110] M.T. Woodside, C. Garcia-Garcia, S.M. Block, Folding and unfolding single RNAmolecules under tension, Curr. Opin. Chem. Biol. 12 (2008) 640–646.

[111] J.E. Barrick, R.R. Breaker, The distributions, mechanisms, and structures ofmetabolite-binding riboswitches, Genome Biol. 8 (2007) R239.

[112] M. Heldal, S. Norland, E.S. Erichsen, R.A. Sandaa, A. Larsen, F. Thingstad, G. Bratbak,Mg2+ as an indicator of nutritional status in marine bacteria, ISME J. 6 (2012)524–530.

[113] M. Meyer, M. Plass, J. Perez-Valle, E. Eyras, J. Vilardell, Deciphering 3'ss selection inthe yeast genome reveals an RNA thermosensor that mediates alternative splicing,Mol. Cell 43 (2011) 1033–1039.

[114] J. Johansson, P. Mandin, A. Renzoni, C. Chiaruttini, M. Springer, P. Cossart, An RNAthermosensor controls expression of virulence genes in Listeriamonocytogenes, Cell110 (2002) 551–561.

[115] M.T. Morita, Y. Tanaka, T.S. Kodama, Y. Kyogoku, H. Yanagi, T. Yura, Translationalinduction of heat shock transcription factor sigma32: evidence for a built-in RNAthermosensor, Genes Dev. 13 (1999) 655–665.

[116] G. Nechooshtan, M. Elgrably-Weiss, A. Sheaffer, E. Westhof, S. Altuvia, A pH-responsive riboregulator, Genes Dev. 23 (2009) 2650–2662.

[117] A. Reining, S. Nozinovic, K. Schlepckow, F. Buhr, B. Furtig, H. Schwalbe, Three-statemechanism couples ligand and temperature sensing in riboswitches, Nature 499(2013) 355–359.

[118] A.L. Edwards, F.E. Reyes, A. Heroux, R.T. Batey, Structural basis for recognition of S-adenosylhomocysteine by riboswitches, RNA 16 (2010) 2144–2155.

[119] K. Bohme, R. Steinmann, J. Kortmann, S. Seekircher, A.K. Heroven, E. Berger, F.Pisano, T. Thiermann, H. Wolf-Watz, F. Narberhaus, P. Dersch, Concerted actionsof a thermo-labile regulator and a unique intergenic RNA thermosensor controlYersinia virulence, PLoS Pathog. 8 (2012) e1002518.

[120] E. Loh, E. Kugelberg, A. Tracy, Q. Zhang, B. Gollan, H. Ewles, R. Chalmers, V. Pelicic, C.M. Tang, Temperature triggers immune evasion by Neisseria meningitidis, Nature502 (2013) 237–240.

[121] Y. Benenson, Synthetic biology with RNA: progress report, Curr. Opin. Chem. Biol.16 (2012) 278–284.

[122] F.J. Isaacs, D.J. Dwyer, J.J. Collins, RNA synthetic biology, Nat. Biotechnol. 24 (2006)545–554.

[123] A. Wittmann, B. Suess, Engineered riboswitches: expanding researchers' toolboxwith synthetic RNA regulators, FEBS Lett. 586 (2012) 2076–2083.

[124] S. Topp, J.P. Gallivan, Emerging applications of riboswitches in chemical biology,ACS Chem. Biol. 5 (2010) 139–148.

[125] N. Dixon, J.N. Duncan, T. Geerlings, M.S. Dunstan, J.E. McCarthy, D. Leys, J.Micklefield, Reengineering orthogonally selective riboswitches, Proc. Natl. Acad.Sci. U. S. A. 107 (2010) 2830–2835.

[126] N. Dixon, C.J. Robinson, T. Geerlings, J.N. Duncan, S.P. Drummond, J. Micklefield,Orthogonal riboswitches for tuneable coexpression in bacteria, Angew. Chem.Int. Ed. Engl. 51 (2012) 3620–3624.

[127] P. Ceres, A.D. Garst, J.G. Marcano-Velazquez, R.T. Batey, Modularity of selectriboswitch expression platforms enables facile engineering of novel genetic regu-latory devices, ACS Synth. Biol. 2 (2013) 463–472.

[128] J.J. Trausch, P. Ceres, F.E. Reyes, R.T. Batey, The structure of a tetrahydrofolate-sensing riboswitch reveals two ligand binding sites in a single aptamer, Structure19 (2011) 1413–1423.

[129] A. Xayaphoummine, V. Viasnoff, S. Harlepp, H. Isambert, Encoding folding paths ofRNA switches, Nucleic Acids Res. 35 (2007) 614–622.

[130] P. Ceres, J.J. Trausch, R.T. Batey, Engineering modular ‘ON’ RNA switches usingbiological components, Nucleic Acids Res. 41 (2013) 10449–10461



























































































































the purine riboswitch as a model system for exploring rna biology and chemistry

Documents