Skip to main content
NCBI home page
RNA logoLink to RNA
. 2008 May;14(5):822–828. doi: 10.1261/rna.988608

The aptamer core of SAM-IV riboswitches mimics the ligand-binding site of SAM-I riboswitches

Zasha Weinberg 1, Elizabeth E Regulski 1, Ming C Hammond 2, Jeffrey E Barrick 2,3, Zizhen Yao 4, Walter L Ruzzo 4,5, Ronald R Breaker 1,2,3
PMCID: PMC2327355  PMID: 18369181

Abstract

A novel family of riboswitches, called SAM-IV, is the fourth distinct set of mRNA elements to be reported that regulate gene expression via direct sensing of S-adenosylmethionine (SAM or AdoMet). SAM-IV riboswitches share conserved nucleotide positions with the previously described SAM-I riboswitches, despite rearranged structures and nucleotide positions with family-specific nucleotide identities. Sequence analysis and molecular recognition experiments suggest that SAM-I and SAM-IV riboswitches share similar ligand binding sites, but have different scaffolds. Our findings support the view that RNA has considerable structural versatility and reveal that riboswitches exploit this potential to expand the scope of RNA in genetic regulation.

Keywords: AdoMet, S-adenosylmethionine, riboswitches, scaffold, Streptomyces

INTRODUCTION

Riboswitches are mRNA elements that specifically bind a particular metabolite and regulate gene expression in response to changing metabolite concentrations (Mandal and Breaker 2004; Winkler and Breaker 2005; Batey 2006; Coppins et al. 2007). S-adenosylmethionine, a cofactor for many methylase enzymes, is recognized by three previously published riboswitch classes: the SAM-I (Epshtein et al. 2003; McDaniel et al. 2003; Winkler et al. 2003), SAM-II (Corbino et al. 2005), and SAM-III (or SMK) (Fuchs et al. 2006) riboswitches. While all three riboswitch families specifically recognize SAM, they have no apparent similarity in sequence or structure.

The discovery of multiple distinct SAM-binding riboswitches, like, for example, the discovery of multiple self-cleaving ribozymes, supports the view that RNA has sufficient structural sophistication to solve the same biochemical challenges in diverse ways. Furthermore, structural studies and sequence analyses of two classes of large ribozymes, group I introns and RNase P RNAs, show that distinct scaffolds in these RNAs support the same positioning of nucleotides in the catalytic core (Michel and Westhof 1990; Krasilnikov et al. 2004; Torres-Larios et al. 2006; Vicens and Cech 2006). If RNAs can easily assume various structures with similar functions, then we expect that the diversity of riboswitch RNA structures could be far greater than what has been found to date.

While searching for new riboswitches in bacteria using bioinformatics (Weinberg et al. 2007; Yao et al. 2007), we identified a conserved, structured RNA motif that was classified as a novel set of SAM-binding riboswitches. These “SAM-IV” riboswitches are predominantly found in Actinomycetales, for example, Mycobacterium tuberculosis (the causative agent of tuberculosis). SAM-IV riboswitches have similarities to the ligand-binding core of SAM-I riboswitches, but have numerous differences elsewhere in architecture and nucleotide identities. Thus, SAM-I and SAM-IV riboswitches might have distinct scaffolds but highly similar ligand-binding cores. Our findings with SAM-IV riboswitches imply that structural mimicry is not restricted to large ribozymes, but may be widespread among diverse kinds of structured RNAs.

RESULTS AND DISCUSSION

Identification of the SAM-IV motif

The SAM-IV motif was found using the CMfinder comparative genomics pipeline (Weinberg et al. 2007; Yao et al. 2007). Most SAM-IV instances are positioned upstream of genes involved in sulfur metabolism and presumably in their 5′ UTRs, which suggests that SAM-IV corresponds to a SAM-binding riboswitch. Almost all known occurrences of SAM-IV RNAs are in the order Actinomycetales (see Supplementary Materials).

The SAM-IV motif core resembles the core of SAM-I

We compared highly conserved regions of SAM-IV with those of SAM-I, SAM-II, and SAM-III riboswitches. SAM-IV shares key sequence and structural features of the SAM-I binding core, but significant differences in nucleotide identities are found elsewhere, and there are considerable deviations in overall architecture (Fig. 1). We defined the core as all nucleotides in the SAM-I three-dimensional (3D) structure (Montange and Batey 2006) having an atom within 5 Å of a ligand atom (Fig. 1, bold letters) and mapped SAM-I positions onto analogous SAM-IV positions based on apparent similarities in conserved structural features. Nucleotide identities were classified as common to both motifs (Fig. 1, yellow shading) when surrounding conserved nucleotide identities or other features suggested to us that the common identity is unlikely to arise by chance.

FIGURE 1.

FIGURE 1.

Open in a new tab

Comparison of SAM-I and SAM-IV motifs. Stems are labeled P1–P5. P5 in SAM-IV is often missing, but its 5′ side involved in the pseudoknot is always present. Nucleotide positions in the published SAM-I 3-D structure (Montange and Batey 2006) within 5 Å of the ligand are depicted in bold, as are their putative corresponding positions in SAM-IV. Six positions proposed to directly contact the ligand (e.g., A45) are labeled in both motifs according to their SAM-I numbering (Montange and Batey 2006). Conserved features (nucleotide identities, bulges, and stems) are shaded to indicate that they are common to both motifs (yellow), unique to SAM-I (pink), or unique to SAM-IV (blue). Covarying mutations support the base pairs in SAM-IV proposed to be conserved. These are summarized in Supplemental Figure S1, and detailed in Supplemental Figure S3D.

Both motifs have a multistem junction and extensive similarity in nucleotide identity in the P3 stem (Fig. 1). The junctions between P1 and P2 also share similarity, and both motifs appear to form a pseudoknot between the terminal loop of P2 and the junction 3′ of P3. To accommodate this pseudoknot, P2 in SAM-I has a canonical kink-turn (Lescoute et al. 2005; Montange and Batey 2006). Interestingly, the internal loop in P2 in SAM-IV might function analogously to the kink turn, allowing these RNAs to adopt a pseudoknot structure.

However, the two motifs have distinct architectural elements and distinct patterns of nucleotide conservation in many places (Fig. 1, pink and blue shading). The P4 hairpin in the SAM-I core is absent in SAM-IV, but a different P4 hairpin is found outside of P1 in SAM-IV. These P4 hairpins have dissimilar conserved nucleotide identities. In addition to the added P4 outside the core, SAM-IV is predicted to form an additional pseudoknot that SAM-I lacks. Also, SAM-IV has a considerably different P2 from SAM-I, in that its loops and stem lengths are different from that of SAM-I, and there appears to be little similarity in sequence. Indeed, the kink-turn in P2 is highly conserved in SAM-I riboswitches (Barrick and Breaker 2007), but no SAM-IV riboswitch sequences have this structural motif. Other differences in conserved nucleotide identities between SAM-I and SAM-IV are in the tip of P3, base of P1, and junction 3′ to P3.

Because of these differences, SAM-IV riboswitches have escaped detection by homology searches based on SAM-I in at least three previous studies. In the first study, Actinobacterial genomes were searched for regulatory RNAs including SAM-I aptamers using the PAT program (Seliverstov et al. 2005). Second, Gram-positive bacteria were searched for SAM-I aptamers and other gene control elements in sulfur metabolism using the RNA-PATTERN program (Rodionov et al. 2004). Third, our searches for homologs of known riboswitches in all bacteria have used SequenceSniffer (J.E. Barrick and R.R. Breaker, unpubl.), and RaveNnA (Weinberg and Ruzzo 2006), which exploits statistical profile techniques to model probabilistically a conserved sequence and secondary structure. All efforts failed to detect SAM-IV riboswitches, although many actinomycetes carry both SAM-I and SAM-IV riboswitches at distinct genome coordinates.

SAM-I and SAM-IV similarities are primarily restricted to their cores, and therefore SAM-I and SAM-IV riboswitches might share similar binding sites and molecular recognition characteristics. This hypothesis is supported by the SAM-I atomic-resolution model (Montange and Batey 2006), in which 6 nucleobases were proposed to interact directly with the ligand (Fig. 1, purple labels). The nucleotide identitities at these six positions are strictly conserved in all but one SAM-I representative. This exceptional SAM-I representative, which is found in Deinococcus radiodurans, fits the consensus poorly in P1 and P4. This representative might be a significant variation of the SAM-I structure or might be a defective riboswitch. Thus, its failure to conserve all six positions might not reflect normal functional constraints on SAM-I riboswitches. Of the six positions in SAM-IV proposed to correspond to these SAM-I ligand-contacting positions, five are conserved in all known SAM-IV representatives with the same nucleotide identity as in SAM-I riboswitches. The sixth conserved nucleobase position in SAM-I is U88. Its corresponding position in SAM-IV is typically a cytosine, but all four nucleobases are observed at this position.

Molecular recognition characteristics of SAM-IV RNAs

To assess molecular recognition of SAM-IV RNAs, we applied in-line probing assays (Soukup and Breaker 1999) to a 132-nucleotide RNA (termed 132 Sc RNA) comprising the SAM-IV found in Streptomyces coelicolor. In these assays, 132 Sc RNA binds SAM with an apparent dissociation constant (K D) of ∼150 nM, versus 20 μM for S-adenosylhomocysteine (SAH). Adenosine and methionine have K D values poorer than 1 mM (Fig. 2). These values demonstrate that the 132 Sc RNA aptamer binds SAM with similar affinity and specificity as other classes of SAM riboswitch aptamers.

FIGURE 2.

FIGURE 2.

Open in a new tab

SAM-IV RNA selectively binds SAM. (A) Sequence and inferred secondary structure of 132 Sc RNA. See Supplemental Figure S2 for additional data. (B) In-line probing gel. (NR) No reaction (RNA only), (T1) partial RNase T1 digest (cleaves 3′ to guanosyl residues), (−OH) partial alkaline digest (cleaves all internucleotide linkages), (−) no compound was added to reaction. SAM, SAH, Ade (adenosine), and Met (methionine) were added as designated. G21–G117: identification of selected bands corresponding to cleavage after G residues. R1–R5: regions undergoing ligand-mediated modulation. (C) Modulating regions (R1–R5) were quantitated from a gel (not shown) with multiple SAM concentrations and normalized to the range 0 (no SAM) to 1 (saturating SAM concentration). The apparent K D is that whose theoretical two-state binding curve best fits the data shown. The theoretical curve for K D = 150 nM is shown.

SAM-IV riboswitches do not conserve the nucleotide corresponding to U88 in SAM-I (Fig. 1). Because the O2 carbonyl oxygen at this position in SAM-I riboswitches is expected to sense the positive charge at the sulfur in the SAM ligand (Montange and Batey 2006), we compared the affinities of SAM-I and SAM-IV riboswitches for compounds that differ at the sulfur position: SAM, SAH, SAH sulfone (Borchardt and Wu 1974), and two aza derivatives, AzaAdoMet and MeAzaAdoMet (Thompson et al. 1999). 132 Sc RNA has a cytosine in place of uracil as the nucleobase in the position analogous to U88. Although cytosine preserves the O2 carbonyl, the additional hydrogen bond formed by a Watson–Crick G-C pair might disturb the interaction with the positive charge of the ligand, which could explain the perfect conservation of U88 in SAM-I riboswitches.

We compared molecular recognition of 132 Sc RNA to that of a previously studied SAM-I riboswitch called 124 yitJ (Winkler et al. 2003). To measure discrimination by a riboswitch against a modification in the ligand, we analyzed the ratio of the K D for the modified ligand divided by the K D for SAM. SAM-I and SAM-IV riboswitches bind SAM and SAM derivatives to yield similar K D ratios (Table 1). These data support the hypothesis that SAM-IV molecular recognition is similar to that of SAM-I, despite the poor conservation of U88 in SAM-IV. It is possible that the U88 position does not strongly interact with the ligand and is conserved in SAM-I for a different purpose. Alternatively, the cytosine in 132 Sc RNA might be able to accomplish the same function within the altered SAM-IV architecture, or a different interaction in SAM-IV might substitute for U88.

TABLE 1.

Molecular recognition by SAM-I and SAM-IV

graphic file with name 822tbl1.jpg

Open in a new tab

SAM-IV is a genetic regulatory element

To determine if SAM-IV regulates gene expression in vivo, we cloned the SAM-IV-containing intergenic region upstream of S. coelicolor locus SCO2146 as a translational fusion with the xylE reporter gene (see Materials and Methods). Because a promoter could not be easily identified in the SAM-IV intergenic region, we placed the region under the control of a glycerol-inducible promoter (Kieser et al. 2000).

Most known SAM riboswitches lower gene expression upon binding SAM, thereby reducing expression of genes whose protein products are not needed when SAM levels are high. To test if SAM-IV is a gene regulation element, we introduced mutations expected to disrupt or restore function of the SAM-IV riboswitch. The first mutation (M1) modified positions expected to directly contact SAM, by assuming that these nucleotides are analogous to those in SAM-I aptamers. The second mutation (M2) disrupts base pairing in the conserved secondary structure. A third mutation (M3) restores base pairing lost in M2 via compensatory mutations, which are predicted to restore wild-type activity.

We measured XylE activity in S. coelicolor cells carrying each of the four riboswitch–reporter fusions after growth in complex media, in which cellular concentrations of SAM are expected to be ample. As anticipated, strains with M1 and M2 riboswitches showed higher reporter gene expression than those with wild-type or M3 riboswitches (Fig. 3). These results indicate that the SAM-IV RNA from S. coelicolor is a genetic control element, which molecular recognition experiments indicate should respond to SAM. We also measured reporter gene expression under conditions in which cellular SAM concentrations are depleted, but these results were inconclusive due to exceedingly low levels of reporter gene expression with the constructs used (see Supplementary Materials).

FIGURE 3.

FIGURE 3.

Open in a new tab

SAM-IV is a genetic regulatory element. (A) The mutations tested (M1, M2, M3) are shown in the aptamer secondary structure, although the entire intergenic region was used in the reporter assay (see Materials and Methods). (B) XylE reporter activity was quantitated as change in A375 per minute (measured over 20 min) per gram of total protein. Cell strains were: (WT) wild-type IGR, (M1–M3) mutated IGR, (no vector) cells lack the vector containing the xylE reporter gene. (C) Absorbance versus time plots for three typical experiments selected from B.

Few SAM-IV aptamers are followed by predicted rho-independent transcription terminators, but many aptamers are near the predicted start of their downstream gene (Supplemental Fig. S3D). Thus, it is most reasonable to infer that ligand binding in most SAM-IV riboswitches leads to sequestration of the downstream ribosome binding site (Barrick and Breaker 2007). Interestingly, some genes apparently regulated by SAM-IV (e.g., SCO2146) are predicted to encode selenocysteine lyases (Supplemental Fig. S3C), so SAM-IV might bind the selenium-derivative of SAM, Se-adenosylselenomethionine (AdoSeMet). However, AdoSeMet has no known biological role, but rather results from excessive selenium levels, and it is normally much less abundant than SAM (Stadtman 1983). Therefore, the predicted selenocysteine lyases regulated by SAM-IV riboswitches might be misannotated. Gene regulation of sulfur metabolism is of particular interest in Streptomyces, as SAM levels affect antibiotic production and sporulation in these species (Kim et al. 2003; Okamoto et al. 2003).

Evolutionary history of SAM-IV

The similarity between SAM-I and SAM-IV is too low to automatically assume a common evolutionary origin. However, we suggest a model that could account for divergent evolution. In this model, a SAM-I-like ancestor loses its P4 stem, and any negative effect later caused by the loss of P4 was compensated by the gain of a different P4 stem 3′ to the P1 stem, leading eventually to a SAM-IV-like RNA. Three facts are consistent with this model. First, SAM-IV is almost restricted to Actinomycetales, while SAM-I is found in this taxon and many others, suggesting that, if they share an ancestor, it would probably resemble SAM-I. Second, P4 was found to be involved only in scaffolding (Montange and Batey 2006), so it might be more easily replaced. Third, a few SAM-I riboswitches entirely lack the P4 stem and loop (see Supplemental Fig. S4D), so P4 loss can be tolerated.

Classification and significance of SAM-IV riboswitches

Proteins are classified into families and superfamilies by varying criteria (Orengo and Thornton 2005). The original criteria were based solely on sequence homology (Dayhoff et al. 1976). Other groups have continued this precedent (Barker et al. 1996), but some also consider structural or functional similarity (Murzin et al. 1995). Generally, proteins classified into families have clear homology, while proteins grouped into superfamilies appear to be more distantly related and homology is considered only “probable” (Dayhoff et al. 1976; Murzin et al. 1995). Given the incompatible patterns of conservation and rearranged structures between SAM-I and SAM-IV riboswitches, and since their similarity eludes state-of-the-art methods of sequence analysis, we propose that they comprise distinct families. However, the similarity of their cores is suggestive enough to classify them as a single superfamily. Three-dimensional structural studies on SAM-IV riboswitches might help to refine their classification. Our laboratory has frequently grouped riboswitches into “classes” and “types.” A class defines a set of riboswitches sharing structural and functional properties, while type is used to group RNAs that belong to the same class, yet carry distinct substructures. To relate this terminology with that used to group homologous proteins, “class” corresponds to “family”, and “type” corresponds to “subfamily.”

Additional riboswitch families show structural diversity, although possibly of a different nature from the relationship between SAM-I and SAM-IV riboswitches. For example, preQ1-I riboswitches can be partitioned into subfamilies based on loop sequences (Roth et al. 2007), but it is unknown how these differences affect riboswitch structure. Also an adenosylcobalamin-binding riboswitch reported previously (Nahvi et al. 2004) lacks a normally conserved domain, while its 3′ flanking sequence is essential for ligand recognition. This variant exhibited better affinity for purinylcobalamin than does a typical adenosylcobalamin riboswitch, so the binding core of the variant might be altered.

Our results demonstrate that riboswitch RNAs can use diverse architectures to position key nucleotides into the same binding site. In view of similar findings for group I introns and RNase P RNAs, we conclude that this phenomenon is exhibited by a wide range of RNAs. This robust natural use of the structural repertoire of RNA supports the conclusion that RNA is a powerful and versatile polymer for molecular sensing and genetic control.

MATERIALS AND METHODS

Bioinformatics

The SAM-IV motif was identified using the CMfinder comparative genomics pipeline, and its alignment inferred using previously established strategies (Weinberg et al. 2007; Yao et al. 2007), but version 21 of RefSeq (Pruitt et al. 2005) was used. SAM-I alignment data were published elsewhere (Barrick and Breaker 2007). Conservation statistics presented in Figure 1 were calculated as previously reported (Weinberg et al. 2007).

Microbial genetics and plasmids

To test riboswitch control of gene expression, we cloned the SAM-IV-containing intergenic region in S. coelicolor upstream of a xylE reporter gene (Kieser et al. 2000) encoding catechol 2,3-dioxygenase. This enzyme converts catechol to 2-hydroxymuconate semialdehyde, whose absorbance at 375 nm permits a colorimetric assay. We used pMT3226 (Kieser et al. 2000), an integrative Streptomyces plasmid that replicates in Escherichia coli and includes an apramycin resistance gene and a xylE reporter gene downstream of a glycerol-inducible promoter. To construct a translation fusion with the xylE gene, a second BamHI site was added to pMT3226 by QuikChange (Stratagene) with primer 5′-GAAGAGGTGACGTCATGGATCCGAACAAAGGTGTAATGC and its reverse complement. The S. coelicolor intergenic region containing SAM-IV was amplified from genomic DNA by PCR with primers 5′-CTAGGATCCGCCACGCGTAGCGGCCCTGGTGTGT and 5′- GTAGGATCCACAGCGGTGGGTACGGACATGGCG (BamHI sites underlined). To aid amplification of the high-G + C DNA product, PCR reactions contained 1.3 M betaine (Sigma-Aldrich) and 5% DMSO (Frackman et al. 1998). To assemble “pEZ35,” pMT3226 and the PCR product were cut with BamHI, the small fragment from pMT3226 was removed by agarose gel purification, and the DNA molecules were ligated, with verification of correct sequence and orientation by sequencing. Vectors containing mutant riboswitches (M1, M2, M3) were made by QuikChange. Plasmids were introduced into S. coelicolor M145 by conjugation (Kieser et al. 2000). Spore stocks of exconjugants were harvested from cells grown on MS agar (Kieser et al. 2000) at 28°C, and stored at −20°C in 15% glycerol. Apramycin sulfate (Sigma-Aldrich) was used at 100 μg/mL for E. coli or 50 μg/mL for S. coelicolor.

To start genetics experiments, spore stocks were thawed on ice, washed in 1 mL of triton X-100 buffer (Hodgson 1982), and resuspended in 1 mL dH2O. This mixture was inoculated into a 25 mL culture to A450 ∼ 0.03, measured relative to H2O (Hodgson 1982). Media was 24 g/L Tryptic Soy Broth (TSB) without dextrose (Becton Dickinson), 4% glycerol, 20 μg/mL apramycin sulfate (except for cells lacking the vector), and 12.5 μg/mL sodium nalidixate (Sigma-Aldrich), to reduce the risk of contamination. Cultures were grown for 50.5 h in 250-mL flasks with shaking at 28°C. Then, XylE activity was measured in cell-free extracts (Kieser et al. 2000). Since cell debris remaining in the crude extract add noise to XylE enzyme curves, the supernatant was extracted after centrifugation twice and was then filtered through a 0.2-μm-pore membrane (Whatman), followed by a 20-min centrifugation. Reaction included 1 mL catechol-free assay buffer, 100 μL cell-free extract, 10 μL 20 mM catechol, and absorbance at 375 nm was measured every 6 sec for 20 min on a Varian 50 Bio spectrophotometer. Despite filtering, debris sometimes remained that absorbed at all visible wavelengths, including 375 nm. When the change at 480 nm (a wavelength not noticeably absorbed by the XylE reaction product) was more than 0.0004 over 20 min and comparable in magnitude to that of 375 nm, we discarded the result and prepared a new reaction. XylE activity was estimated using the change in A375 over the full 20 min, which we expect to be more reproducible than the use of just the linear range in an earlier protocol (Kieser et al. 2000).

In-line probing assays

DNA corresponding to the aptamer in the pEZ35 plasmid (above) was amplified by PCR with primers 5′-TAATACGACTCACTATAGGTTTTTCGACAGGTCATGAGTGACAGTC (T7 RNA polymerase promoter sequence underlined) and 5′-AGGGGTCCGCGCTTGCCGTGGACCTTGCTG. 5′ 32P-labeled RNA molecules were prepared from PCR products by in vitro transcription, dephosphorylation, and radiolabeling with [γ-32P]ATP using protocols described previously (Roth et al. 2007). In-line probing reactions were prepared, and 5′ 32P-labeled fragments were resolved by denaturing PAGE and imaged as described previously (Roth et al. 2007).

SUPPLEMENTAL DATA

Supplemental material can be found at http://www.rnajournal.org.

ACKNOWLEDGMENTS

We thank Mark Buttner, Maureen Bibb, David Hopwood, and Stanley Cohen for supplying vectors and strains for the Streptomyces work, G. Michael Blackburn for aza derivatives of SAM, and JinSoo Lim for SAH sulfone. We also thank Nicholas Carriero and Robert Bjornson for assisting our use of the Yale Life Sciences High Performance Computing Center (NIH grant RR19895-02), Shane Neph and Martin Tompa for bioinformatics work assisting motif identification, and Michael Vockenhuber, N. Sudarsan, Mark Buttner, Ralph Bertram, Matt Cabeen, and Breaker laboratory members for helpful advice. This work was funded by the NIH (GM 068819, R01 HG02602), NSF (EIA-0323510, DBI-0218798), and the Howard Hughes Medical Institute (HHMI). J.E.B. was a HHMI predoctoral fellow.

Footnotes

Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.988608.

REFERENCES

  1. Barker, W.C., Pfeiffer, F., George, D.G. Superfamily classification in PIR-International Protein Sequence Database. Methods Enzymol. 1996;266:59–71. doi: 10.1016/s0076-6879(96)66006-6. [DOI] [PubMed] [Google Scholar]
  2. Barrick, J.E., Breaker, R.R. The distributions, mechanisms, and structures of metabolite-binding riboswitches. Genome Biol. 2007;8:R239. doi: 10.1186/gb-2007-8-11-r239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Batey, R.T. Structures of regulatory elements in mRNAs. Curr. Opin. Struct. Biol. 2006;16:299–306. doi: 10.1016/j.sbi.2006.05.001. [DOI] [PubMed] [Google Scholar]
  4. Borchardt, R.T., Wu, Y.S. Potential inhibitors of S-adenosylmethionine-dependent methyltransferases. 1. Modification of the amino acid portion of S-adenosylhomocysteine. J. Med. Chem. 1974;17:862–868. doi: 10.1021/jm00254a016. [DOI] [PubMed] [Google Scholar]
  5. Coppins, R.L., Hall, K.B., Groisman, E.A. The intricate world of riboswitches. Curr. Opin. Microbiol. 2007;10:176–181. doi: 10.1016/j.mib.2007.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Corbino, K.A., Barrick, J.E., Lim, J., Welz, R., Tucker, B.J., Puskarz, I., Mandal, M., Rudnick, N.D., Breaker, R.R. Evidence for a second class of S-adenosylmethionine riboswitches and other regulatory RNA motifs in α-proteobacteria. Genome Biol. 2005;6:R70. doi: 10.1186/gb-2005-6-8-r70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dayhoff, M.O., Barker, W.C., Hunt, L.T. Protein superfamilies. In: Dayhoff M.O., editor. Atlas of protein sequence and structure. Suppl. 2. Vol. 5. National Biomedical Research Foundation; Washington, DC: 1976. pp. 9–19. [Google Scholar]
  8. Epshtein, V., Mironov, A.S., Nudler, E. The riboswitch-mediated control of sulfur metabolism in bacteria. Proc. Natl. Acad. Sci. 2003;100:5052–5056. doi: 10.1073/pnas.0531307100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Frackman, S., Kobs, G., Simpson, D., Storts, D. Betaine and DMSO: Enhancing agents for PCR. Promega Notes. 1998;65:27. [Google Scholar]
  10. Fuchs, R.T., Grundy, F.J., Henkin, T.M. The S(MK) box is a new SAM-binding RNA for translational regulation of SAM synthetase. Nat. Struct. Mol. Biol. 2006;13:226–233. doi: 10.1038/nsmb1059. [DOI] [PubMed] [Google Scholar]
  11. Hodgson, D.A. Glucose repression of carbon source uptake in Streptomyces coelicolor A3(2) and its perturbation in mutants resistant to 2-deoxyglucose. J. Gen. Microbiol. 1982;128:2417–2430. [Google Scholar]
  12. Kieser, T., Bibb, M.J., Buttner, M.J., Chater, K.F., Hopwood, D.A. The John Innes Foundation; Norwich, UK: 2000. Practical Streptomyces genetics . [Google Scholar]
  13. Kim, D.J., Huh, J.H., Yang, Y.Y., Kang, C.M., Lee, I.H., Hyun, C.G., Hong, S.K., Suh, J.W. Accumulation of S-adenosyl-L-methionine enhances production of actinorhodin but inhibits sporulation in Streptomyces lividans TK23. J. Bacteriol. 2003;185:592–600. doi: 10.1128/JB.185.2.592-600.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Krasilnikov, A.S., Xiao, Y., Pan, T., Mondragon, A. Basis for structural diversity in homologous RNAs. Science. 2004;306:104–107. doi: 10.1126/science.1101489. [DOI] [PubMed] [Google Scholar]
  15. Lescoute, A., Leontis, N.B., Massire, C., Westhof, E. Recurrent structural RNA motifs, isostericity matrices and sequence alignments. Nucleic Acids Res. 2005;33:2395–2409. doi: 10.1093/nar/gki535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lim, J., Winkler, W.C., Nakamura, S., Scott, V., Breaker, R.R. Molecular-recognition characteristics of SAM-binding riboswitches. Angew. Chem. Int. Ed. Engl. 2006;45:964–968. doi: 10.1002/anie.200503198. [DOI] [PubMed] [Google Scholar]
  17. Mandal, M., Breaker, R.R. Gene regulation by riboswitches. Nat. Rev. Mol. Cell Biol. 2004;5:451–463. doi: 10.1038/nrm1403. [DOI] [PubMed] [Google Scholar]
  18. McDaniel, B.A., Grundy, F.J., Artsimovitch, I., Henkin, T.M. Transcription termination control of the S box system: Direct measurement of S-adenosylmethionine by the leader RNA. Proc. Natl. Acad. Sci. 2003;100:3083–3088. doi: 10.1073/pnas.0630422100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Michel, F., Westhof, E. Modeling of the three-dimensional architecture of group I catalytic introns based on comparative sequence analysis. J. Mol. Biol. 1990;216:585–610. doi: 10.1016/0022-2836(90)90386-Z. [DOI] [PubMed] [Google Scholar]
  20. Montange, R.K., Batey, R.T. Structure of the S-adenosylmethionine riboswitch regulatory mRNA element. Nature. 2006;441:1172–1175. doi: 10.1038/nature04819. [DOI] [PubMed] [Google Scholar]
  21. Murzin, A.G., Brenner, S.E., Hubbard, T., Chothia, C. SCOP: A structural classification of proteins database for the investigation of sequences and structures. J. Mol. Biol. 1995;247:536–540. doi: 10.1006/jmbi.1995.0159. [DOI] [PubMed] [Google Scholar]
  22. Nahvi, A., Barrick, J.E., Breaker, R.R. Coenzyme B12 riboswitches are widespread genetic control elements in prokaryotes. Nucleic Acids Res. 2004;32:143–150. doi: 10.1093/nar/gkh167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Okamoto, S., Lezhava, A., Hosaka, T., Okamoto-Hosoya, Y., Ochi, K. Enhanced expression of S-adenosylmethionine synthetase causes overproduction of actinorhodin in Streptomyces coelicolor A3(2) J. Bacteriol. 2003;185:601–609. doi: 10.1128/JB.185.2.601-609.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Orengo, C.A., Thornton, J.M. Protein families and their evolution—A structural perspective. Annu. Rev. Biochem. 2005;74:867–900. doi: 10.1146/annurev.biochem.74.082803.133029. [DOI] [PubMed] [Google Scholar]
  25. Pruitt, K., Tatusova, T., Maglott, D. NCBI Reference Sequence (RefSeq): A curated nonredundant sequence database of genomes, transcripts, and proteins. Nucleic Acids Res. 2005;33:D501–D504. doi: 10.1093/nar/gki011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rodionov, D.A., Vitreschak, A.G., Mironov, A.A., Gelfand, M.S. Comparative genomics of the methionine metabolism in Gram-positive bacteria: A variety of regulatory systems. Nucleic Acids Res. 2004;32:3340–3353. doi: 10.1093/nar/gkh659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Roth, A., Winkler, W.C., Regulski, E.E., Lee, B.W., Lim, J., Jona, I., Barrick, J.E., Ritwik, A., Kim, J.N., Welz, R., et al. A riboswitch selective for the queuosine precursor preQ1 contains an unusually small aptamer domain. Nat. Struct. Mol. Biol. 2007;14:308–317. doi: 10.1038/nsmb1224. [DOI] [PubMed] [Google Scholar]
  28. Seliverstov, A.V., Putzer, H., Gelfand, M.S., Lyubetsky, V.A. Comparative analysis of RNA regulatory elements of amino acid metabolism genes in Actinobacteria. BMC Microbiol. 2005;5:54. doi: 10.1186/1471-2180-5-54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Soukup, G.A., Breaker, R.R. Relationship between internucleotide linkage geometry and the stability of RNA. RNA. 1999;5:1308–1325. doi: 10.1017/s1355838299990891. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Stadtman, T.C. New biologic functions—Selenium-dependent nucleic acids and proteins. Fundam. Appl. Toxicol. 1983;3:420–423. doi: 10.1016/s0272-0590(83)80015-3. [DOI] [PubMed] [Google Scholar]
  31. Thompson, M.J., Mekhalfia, A.D.P.H., Blackburn, G.M. Synthesis of two stable nitrogen analogues of S-adenosyl-L-methionine. J. Org. Chem. 1999;64:7467–7473. [Google Scholar]
  32. Torres-Larios, A., Swinger, K.K., Pan, T., Mondragon, A. Structure of ribonuclease P-a universal ribozyme. Curr. Opin. Struct. Biol. 2006;16:327–335. doi: 10.1016/j.sbi.2006.04.002. [DOI] [PubMed] [Google Scholar]
  33. Vicens, Q., Cech, T.R. Atomic level architecture of group I introns revealed. Trends Biochem. Sci. 2006;31:41–51. doi: 10.1016/j.tibs.2005.11.008. [DOI] [PubMed] [Google Scholar]
  34. Weinberg, Z., Ruzzo, W.L. Sequence-based heuristics for faster annotation of non-coding RNA families. Bioinformatics. 2006;22:35–39. doi: 10.1093/bioinformatics/bti743. [DOI] [PubMed] [Google Scholar]
  35. Weinberg, Z., Barrick, J.E., Yao, Z., Roth, A., Kim, J.N., Gore, J., Wang, J.X., Lee, E.R., Block, K.F., Sudarsan, N., et al. Identification of 22 candidate structured RNAs in bacteria using the CMfinder comparative genomics pipeline. Nucleic Acids Res. 2007;35:4809–4819. doi: 10.1093/nar/gkm487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Winkler, W.C., Breaker, R.R. Regulation of bacterial gene expression by riboswitches. Annu. Rev. Microbiol. 2005;59:487–517. doi: 10.1146/annurev.micro.59.030804.121336. [DOI] [PubMed] [Google Scholar]
  37. Winkler, W.C., Nahvi, A., Sudarsan, N., Barrick, J.E., Breaker, R.R. An mRNA structure that controls gene expression by binding S-adenosylmethionine. Nat. Struct. Biol. 2003;10:701–707. doi: 10.1038/nsb967. [DOI] [PubMed] [Google Scholar]
  38. Yao, Z., Barrick, J., Weinberg, Z., Neph, S., Breaker, R., Tompa, M., Ruzzo, W.L. A computational pipeline for high-throughput discovery of cis-regulatory noncoding RNA in prokaryotes. PLoS Comput. Biol. 2007;3:e126. doi: 10.1371/journal.pcbi.0030126. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from RNA are provided here courtesy of The RNA Society

RESOURCES