Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
GLMS RIBOZYME COENZYMES AND METHODS OF USE THEREOF
Document Type and Number:
WIPO Patent Application WO/2006/133215
Kind Code:
A2
Abstract:
The present invention provides glmS ribozyme coenzymes and methods of use thereof.

Inventors:
SOUKUP GARRETT A (US)
SOUKUP JULIANE K (US)
Application Number:
US2006/021965
Publication Date:
December 14, 2006
Filing Date:
June 06, 2006
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV CREIGHTON (US)
SOUKUP GARRETT A (US)
SOUKUP JULIANE K (US)
International Classes:
C12Q1/25; C12Q1/68
Foreign References:
US20030040114A12003-02-27
Other References:
WINKLER WADE C ET AL: "Control of gene expression by a natural metabolite-responsive ribozyme" NATURE; NATURE MAR 18 2004, vol. 428, no. 6980, 18 March 2004 (2004-03-18), pages 281-286, XP002410184
KNUDSEN SCOTT M ET AL: "Ribozyme déjà vu." NATURE STRUCTURAL & MOLECULAR BIOLOGY. APR 2004, vol. 11, no. 4, April 2004 (2004-04), pages 301-303, XP002410185 ISSN: 1545-9993
JENNE A ET AL: "Rapid identification and characterization of hammerhead-ribozyme inhibitors using fluorescence-based technology" NATURE BIOTECHNOLOGY, NATURE PUBLISHING GROUP, NEW YORK, NY, US, vol. 19, no. 1, January 2001 (2001-01), pages 56-61, XP002225624 ISSN: 1087-0156
MOROHOSHI FUMIKO ET AL: "Diverse capacities for the adaptive response to DNA alkylation in Bacillus species and strains" MUTATION RESEARCH, vol. 337, no. 2, 1995, pages 97-110, XP008072354 ISSN: 0027-5107
BREAKER R R ET AL: "Riboswitch RNAs: A new class of antimicrobial targets." ABSTRACTS OF THE INTERSCIENCE CONFERENCE ON ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, vol. 43, 2003, pages 94-95, XP008072612 & 43RD ANNUAL INTERSCIENCE CONFERENCE ON ANTIMICROBIAL AGENTS AND CHEMOTHERAPY; CHICAGO, IL, USA; SEPTEMBER 14-17, 2003
MCCARTHY ET AL: "Ligand Requirements for glmS Ribozyme Self-Cleavage" CHEMISTRY AND BIOLOGY, CURRENT BIOLOGY, LONDON, GB, vol. 12, no. 11, November 2005 (2005-11), pages 1221-1226, XP005170410 ISSN: 1074-5521
Attorney, Agent or Firm:
NETTER, Robert, C., Jr. et al. (DANN DORFMAN HERRELL & SKILLMAN, 1601 Market Street Suite 240, Philadelphia PA, 19103-2307, US)
Download PDF:
Claims:

What is claimed is:

1. A method for identifying modulators of a glmS ribozyme, said method comprising: a) contacting said glmS ribozyme with at least one test compound; and b) determining the amount of cleavage by said glmS ribozyme; wherein a change in the amount of cleavage detected in the presence of said at least one compound indicates that said at least one test compound is a modulator of said glmS ribozyme activity.

2. The method of claim 1, wherein said glmS ribozyme is detectably labelled.

3. The method of claim 1, wherein said at least one test compound increases the amount of cleavage.

4. The method of claim 1, wherein said at least one test compound decreases the amount of cleavage.

5. The method of claim 1, wherein said glmS ribozyme is from a Gram positive bacteria.

6. The method of claim 5, wherein said glmS ribozyme is from B. cereus.

7. The method of claim 1, wherein said method is performed in vitro.

8. The method of claim 1, wherein said method is performed in a cell .

9. The method of claim 1, further comprising the step of determining the antibacterial properties of said at least one test compound.

10. A method for modulating the cleavage activity of a glmS ribozyme comprising contacting said glmS ribozyme with a coenzyme, wherein said coenzyme is not D- glucosamine-β-phosphate (GlcNβP) .

11. The method of claim 10, wherein said coenzyme is selected from the group consisting of GIcN, serinol, and GlcNβP analogs.

12. The method of claim 11, wherein said GlcNβP analog is selected from the group consisting of phosphonate analogs and halogenated phosphonate analogs .

13. The method of claim 10, wherein said glmS ribozyme is operably linked to a nucleic acid molecule of interest.

14. The method of claim 13, wherein the modulation of the cleavage of said glmS ribozyme modulates the expression of the nucleic acid molecule of interest.

15. A kit for the modulation of the cleavage activity of a glmS ribozyme comprising a) at least one nucleic acid molecule comprising said glmS ribozyme; and b) at least one coenzyme, with the proviso that at least one of the coenzymes is not GlcNβP.

16. The kit of claim 15, wherein said nucleic acid

molecule is a vector.

17. The kit of claim 15, wherein said coenzyme is selected from the group consisting of GIcN, serinol, and GlcNβP analogs .

18. The kit of claim 15, wherein said GlcNβP analog is selected from the group consisting of phosphonate analogs and halogenated phosphonate analogs .

Description:

glmS Ribozyme Coenzymes and Methods of Use Thereof

By Garrett A. Soukup

Juliane K. Soukup

This application claims priority under 35

U. S. C. §119 (e) to U.S. Provisional Patent Application No. 60/687,717, filed on June 6, 2005. The foregoing application is incorporated by reference herein.

Pursuant to 35 U. S. C. Section 202 (c), it is acknowledged that the United States Government has certain rights in the invention described herein, which was made in part with funds from the National Institutes of Health Grant No. P20 RR16469.

FIELD OF THE INVENTION

The present invention relates to glmS ribozymes, coenzymes thereof, and methods of use thereof.

BACKGROUND OF THE INVENTION

Natural RNA catalysts (ribozymes) perform essential reactions in biological RNA processing (Frank and Pace (1998) Annu. Rev. Biochem. , 67:153-180; Doudna and Cech (2002) Nature, 418:222-228) and protein synthesis (Moore and Steitz (2003) Annu. Rev. Biochem., 72:813-850), whereby catalysis is directly attributable to RNA structure alone or in combination with metal ion cofactors. The majority of natural ribozymes perform RNA cleavage or splicing through transesterification reactions that utilize an internal or external hydroxyl group for nucleophilic attack on the phosphate at the cleavage or splice site (Moore and Steitz (2003) Annu. Rev. Biochem., 72:813-850). Mechanistic strategies that ribozymes employ to promote RNA transesterification include metal-ion catalysis and/or general acid-base

catalysis utilizing internal functional groups, as indicated by the self-cleaving hammerhead, HDV, hairpin, and VS ribozymes (Murray et al . (1998) Chem. Biol., 5:587-595; O'Rear et al . (2001) RNA, 7:537-545; Curtis and Bartel (2001) RNA, 7:546-552; Ferre-D' Amare and Doudna (1998) Nature, 395:567-574; Nakano et al . (2000) Science, 287:1492-1497; Shih and Been (2001) Proc. Natl. Acad. Sci. U.S.A., 98:1489-1494; Rupert and Ferre- D'Amare (2001) Nature, 410:780-786; Rupert et al . (2002) Science, 298:1421-1424; Bevilacqua, P. C. (2003)

Biochemistry, 42:2259-2265; Jones and Strobel (2003) Biochemistry, 42:4265-4276; Lilley, D.M. (2004) RNA, 10:151-158). Although self-splicing group I introns utilize exogenous guanosine as, a nucleophilic substrate (Moore and Steitz (2003) Annu. Rev. Biochem. , 72:813- 850) , catalysis is likewise attributable to intrinsic RNA structure and metal ion cofactors (Adams et al . (2004) Nature, 430:45-50). To date, no natural RNA catalyst has been demonstrated to exploit an exogenous organic compound as a coenzyme for activity, a fundamental strategy that greatly enhances the proficiency of protein enzymes in biological catalysis. The ability of nucleic acid catalysts to utilize coenzymes has only been demonstrated previously by artificial activities (Jadhav and Yarus (2002)

Biochemie, 84:877-888; Roth and Breaker (1998) Proc. Natl. Acad. Sci. U.S.A., 95:6027-6031).

Recent discoveries of two ligand-responsive self- cleaving ribozymes (Winkler et al . (2004) Nature, 428:281-286; Teixeira et al . (2004) Nature, 432:526-530) suggest that the scope of biological RNA catalysis might extend well beyond the intrinsic capabilities of RNA by harnessing cofactor functionalities.

The co-transcriptional cleavage (CoTC) element in the 3 '-untranslated region (UTR) of human β-globin mRNA contains a guanosine-sensitive ribozyme that promotes self-cleavage (Teixeira et al . (2004) Nature, 432:526- 530) and exonuclease-mediated transcriptional termination (West et al . (2004) Nature, 432:522-525). The products of β-globin ribozyme cleavage possess 5'- phosphate and 3' -hydroxy1 termini that do not incorporate guanosine (Teixeira et al . (2004) Nature 432:526-530) .

The glmS ribozyme (Winkler et al . (2004) Nature, 428:281-286) represents a unique member of otherwise non-catalytic 'riboswitcb.es' that function as metabolite-dependent genetic regulatory elements (Mandal and Breaker (2004) Nat. Rev. MoI. Cell Biol., 5:451- 463) . The glmS ribozyme resides in the 5'-UTR of glmS mRNA in certain prokaryotes (Barrick et al . (2004) Proc . Natl. Acad. Sci . U.S.A., 101:6421-6426) and represses gene expression when activated by glucosamine-6- phosphate (GlcNδP) , the metabolic product of the GlmS enzyme (Winkler et al . (2004) Nature, 428:281-286). In other words, the glmS ribozyme functions as a glucosamine-6-phosphate (GlcNδP) -dependent λ riboswitch' to regulate amino-sugar biosynthesis in certain prokaryotes (Winkler et al . (2004) Nature, 428:281-286). The products of glmS ribozyme cleavage possess 5'- hydroxyl and 2 ',3 '-cyclic phosphate termini that do not incorporate GlcNδP (Winkler et al . (2004) Nature, 428:281-286) . While both the β-globin and glmS ribozymes are activated by binding non-substrate organic compounds, it is presently unclear whether their ligands function as effectors in a manner similar to engineered allosteric ribozymes (Silverman, S. K. (2003) RNA, 9:377-383), or as

active participants (i.e. coenzymes) in mechanisms of RNA catalysis .

SUMMARY OF THE INVENTION In accordance with the present invention, methods for identifying modulators of a glmS ribozyme are provided. In one embodiment, the method comprises contacting a glmS ribozyme with at least one test compound and determining the amount of cleavage by the glmS ribozyme, wherein a change in the amount of cleavage indicates that the at least one compound is a modulator of the glmS ribozyme. In another embodiment, the screening method further comprises determining the antibacterial properties of the at least one compound. In another aspect of the instant invention, methods for modulating the cleavage activity of a glmS ribozyme are provided. In one embodiment, the method comprises contacting the glmS ribozyme with a coenzyme which is not the natural coenzyme D-glucosamine-6-phosphate (GlcNβP) . In a particular embodiment, the coenzyme is selected from the group consisting of glucosamine (GIcN), serinol, and GlcNβP analogs such as, without limitation, phosphonate analogs and halogenated phosphonate analogs. In yet another embodiment, the glmS ribozyme is operably linked to a nucleic acid of interest such that the modulation of the cleavage of the glmS ribozyme modulates the expression of the nucleic acid molecule of interest.

According to another aspect of the instant invention, kits are provided for the modulation of the cleavage activity of a glmS ribozyme. In a particular embodiment the kits comprise at least one nucleic acid molecule comprising said glmS ribozyme and at least one coenzyme, wherein at least one of the coenyzmes is not

GlcNβP .

BRIEF DESCRIPTIONS OP THE DRAWING

Figure IA is a refined secondary structure model of the B. cereus glmS ribozyme containing nucleotides -13 through +141 relative to the cleavage site (A) (SEQ ID NO: 1). An additional 5' terminal guanosine at -14 (parentheses) was included to facilitate in vitro transcription. Nucleotide identities that are ≥ 95% conserved are highlighted. Figures IB and 1C provide images of gels of the products of self-cleavage reactions of glmS ribozyme and P1-P3 glmS ribozyme respectively, incubated in the absence (-) or presence (+) of 200 μM GlcNβP (L) under standard conditions (HEPES) or standard conditions substituting HEPES with TRIS (TRIS) . Bands corresponding to the ribozyme (open triangle) and its 3 ' cleavage product (closed triangle) are indicated. Figure ID is a graphical representation of the fraction of glmS ribozyme cleaved in reactions incubated under standard conditions replacing HEPES with HEPES: TRIS in millimolar ratios from 49:1 to 0:50 as indicated by TRIS concentration alone. Error bars indicate the standard deviation determined for two replicate assays . Figure IE is a graphical representation of the rate of glmS ribozyme self- cleavage in the absence (o) or presence (•) of 10 mM TRIS as ligand under standard conditions.

Figure 2A provides the chemical structures of GlcNβP and related compounds depicted as the predominant species at pH 7.5. The portion of GlcNβP mimicked by other amine-containing compounds is shaded. Shown in brackets is the pKa of each amine functionality taken from product literature or experimentally determined. Figure 2B provides an image of a gel of

reaction products of ligand-activated glmS ribozyme self-cleavage incubated under standard conditions in the absence (-) or presence of 200 μM GlcNβP or 10 iriM analog as indicated. Bands corresponding to the ribozyme (open triangle) and its 3' cleavage product (closed triangle) are indicated. Figure 2C provides a graphical representation of the fraction of gliaS ribozyme cleaved in reactions incubated under standard conditions including 10 mM ligand and 0, 1, or 10 mM GlcβP as indicated. Fraction cleaved was normalized to reaction lacking ligand, and error bars indicate the standard deviation determined for two replicate assays. Figure 2D provides a graphical representation of the competition of GlcNβP-activated ribozyme activity by GlcβP. lOμM GlcNβP was incubated with various concentrations of GlcβP. The concentration of GlcβP required to achieve a half maximal K Obs value (apparent Ki) is indicated. Error bars indicate the standard deviation determined for three replicate assays . Figure 2E provides the structures of GlcNβP and certain analogs .

Figure 3A is a graphical representation of reactions performed under standard conditions with various concentrations of GIcN (o) or serinol (δ) . Dashed lines represent a slope of 1. Figure 3B is a graphical representation of reactions performed under standard conditions using 10 mM GIcN (o) or serinol (δ) , and substituting HEPES with other non-amine-containing buffering agents as appropriate to maintain the desired pH. Reactions were similarly performed using 10 mM GlcNβP (•) without added MgCl 2 . The pH required to achieve a half-maximal k ohs value is indicated for each ligand. Data depicted are representative of two replicate assays . Figure 3C is a graphical

representation of reactions performed under standard conditions at pH 7.5 (•) or at pH 8.5 (A) with various concentrations of GlcNβP. Figure 3D is a graphical representation of the pH dependence of the apparent K D for GlcNβP at various pH. Figure 3E is a graphical representation of the GlcNβP concentration dependence of the apparent pKa. Error bars indicate the standard deviation determined for two replicate assays. Figure 3F consists of graphs of the pH-dependence of the apparent K D for GlcNβP. Data represent the average of two replicate assays and error bars indicate standard deviation. Figure 3G consists of graphs of the GlcNβP concentration-dependence of apparent pKa. Data represent the average of two replicate assays, and error bars indicate standard deviation.

Figures 4A and 4B provide graphical representations of the results for the determination of pX a values for GlcNβP, GIcN, and serinol . Figure 4A provides potentiometric titration curves for 10 mM serinol (A) and GIcN (B) and the reciprocal of the change in pH per equivalent of added NaOH (1/δpH) for serinol (C) and GIcN (D) . Figure 4B provides potentiometric titration curves for 1 mM GlcNβP in the presence or absence of 1 mM MgCl 2 (A and B, respectively) and the reciprocal of the change in pH per equivalent of added NaOH (1/δpH) in the presence (C) and absence (D) of 1 mM MgCl 2 . Approximate pX a values are indicated.

Figures 5A and 5B are graphical representations of the GlcNβP-activated glmS ribozyme self-cleavage reaction under varying concentrations of GlcNβP (Fig. 5A) and varying EDTA concentrations (Fig. 5B) .

Figure 6 provides a graphical representation of the observed rate constant (k O bs) versus magnesium ion

concentration for the P1-P3 (open squares) and P1-P4 (filled squares) ribozymes .

Figures 7A and 7B are images of autoradiograms depicting examples of phosphorothioate nucleotide analog (NaS) incorporation in unreacted (uncleaved) and reacted (cleaved) P1-P3 ribozyme (Fig. 7A) and P1-P4 ribozyme (Fig. 7B) . Ribozymes were reacted with GlcNβP under selective conditions in the presence of magnesium ions or manganese ions as indicated. Nucleotide positions of interest are indicated to the left of each gel. Sites of phosphorothioate interference are denoted at the left of corresponding bands with a filled circle. Sites of manganese rescue are denoted at the left of corresponding bands with an asterisk. Identification of positions at the 5 ' ends of reacted ribozymes was aided by the inclusion of hydroxide cleavage ladders (OH) . Figures 8A and 8B are histograms depicting the backbone analog interference and rescue effects in the P1-P3 and P1-P4 B. cereus glmS ribozymes. Phosphorothioate and 2'deoxy interference effects are shown in Figures 8A and 8B, respectively. The left histograms are the P1-P3 ribozyme and the right histograms are the P1-P4 ribozyme. The histograms depict the magnitude of interference (K value) versus nucleotide position superimposed on the secondary structure model for each ribozyme. K values <1 indicate no interference, whereas K values in the range of 2.0 to 4.0, 4.0 to 6.0, and >6.0 are defined as weak, moderate, and strong interferences, respectively. Only K values determined to be > 2.0 are shown, and K values >6.0 are assigned a value of 6.0. Bars correspond to the scale at left and represent average K values from at least two independent experiments conducted with magnesium ions. Numbers represent nucleotide positions within the B.

cereus glmS ribozyme . Sites of phosphorothioate interference that exhibit partial or full manganese rescue are denoted with a single or double asterisk, respectively: Sites of 2 '-deoxy interference that are absent from experiments conducted with manganese ions are indicated by an asterisk. Gray boxes denote Gl of the glmS ribozyme.

Figures 9A and 9C provide images of autoradiograms depicting examples of phosphorothioate nucleotide analog (NaS) incorporation in unreacted (uncleaved) and reacted (cleaved) ribozymes in the presence of magnesium ions. Nucleotide positions that exhibit phosphorothioate interference with GlcNβP are indicated adjacent to corresponding bands at the right of each gel. Sites of partial interference suppression and full interference suppression are denoted by daggers and double daggers, respectively. Identification of positions at the 5' ends of ribozymes was aided by the inclusion of hydroxide cleavage ladders (OH) . Figures 9B and 9D provide graphical representations of the phosphorothioate interferences shown in Figures 9A and 9C. Depicted is the magnitude of interference (K value) for each site. K values represent the average from at least two independent experiments. Figures 1OA and 1OB are histograms of adenosine analog interference and guanosine-analog interference suppression effects, respectively, on P1-P4 B. cereus glmS ribozyme with GIcN and GlcNβP. Histograms depict the magnitude of interference (K value) as described in the legend to Figure 8. Sites of analog interference with GlcNβP that exhibit partial or full suppression with GIcN are denoted with a single or double dagger, respectively. Experiments performed with IaS and

7dGαS did not exhibit GlcN-dependent interference suppression.

Figure 11 provides a model which summarizes backbone and nucleobase functional groups indicated by interference suppression to compose the ligand phosphate-binding site within the P1-P4 ribozyme. Nucleotides within Jl/2 and Pl .1 converge upon the phosphate moiety of GlcNβP (encircled) , where functional group interactions include three phosphate oxygens (PO) , a 2' hydroxyl (OH), and nucleobase imine (N7) and amine (N6) functionalities that are likely metal ion- or water-mediated. Steric boundaries within the phosphate recognition site are indicated at two adenosine positions (C2) .

DETAILED DESCRIPTION OF THE INVENTION

Herein, it is demonstrated that GlcNβP is a coenzyme in glmS ribozyme self-cleavage. Data is provided showing that GlcNβP and other ligands play a role in general acid-base catalysis of RNA transesterification and, therefore, illustrate an expanded capacity for biological RNA catalysis. It is also demonstrated that glmS ribozyme self-cleavage activity both requires and is dependent upon the τpK a (where K 3. is the acid dissociation constant) of the amine functionality of GlcNβP and related compounds, consistent with a role for the ligand in general acid- base catalysis.

Figure IA depicts paired regions P1-P4 of the glmS ribozyme from Bacillus cereus including a putative 5'- proximal pseudoknot (PIa) supported by mutation analysis and a 3 '-proximal pseudoknot (P3a) indicated by phylogenetic analysis. The glmS ribozyme contains four paired domains (P1-P4) that are highly conserved among

an 18-member phylogeny (Barrick et al . (2004) Proc . Natl. Acad. Sci . U.S.A., 101:6421-6426). Previous characterization of the B. subtilis glmS ribozyme demonstrated that the minimal segment required for establishing GlcN6P-dependent self-cleavage activity encompasses nucleotides from the +1 position (relative to the cleavage site) through the P2 domain, while the P3-P4 domains serve to enhance activity (Winkler et al . (2004) Nature, 428:281-286). For the full-length B. subtilis glmS ribozyme, GlcNβP binding was reported to elicit a 1000-fold increase in the observed rate constant (.k o b s ) for self-cleavage activity under saturating GlcNδP and divalent metal ion concentrations (Winkler et al . (2004) Nature, 428:281-286). However, the instant analysis of the B. cereus glmS ribozyme reveals that GlcN6P is more vital to self-cleavage activity than previously appreciated.

In accordance with one aspect of the instant invention, a method is provided for identifying modulators of the glmS ribozyme. Such modulators may increase or decrease the self-cleaving activity of glmS ribozyme. The glmS ribozyme may be optionally detectably labeled with, for example, a fluorescent label or radiolabel. Methods of detectably labeling a nucleic acid molecule are well known in the art (see, e.g., Ausubel et al . , eds . (1998) Current Protocols in Molecular Biology, John Wiley and Sons, Inc.) . Alternatively, the glmS ribozyme and cleavage products thereof may be detected with, for example, a detectably labeled probe. In an exemplary screening method of the instant invention, at least one test compound is incubated with the optionally detectably labeled glmS ribozyme or derivative thereof and the amount of cleavage is determined. A modulation in the amount of

cleavage indicates that the at least one test compound is capable of modulating the self-cleaving activity of the glmS ribozyme .

In another embodiment of the instant invention, the ability of one or more test compounds to modulate glmS ribozyme self-cleaving activity can also be determined in the presence of a ligand or coenzyme known to modulate glmS ribozyme activity, such as GlcNδP. For example, the instant method can be employed to identify test compounds capable of competing with, for example, GlcNβP to effect modulation of glmS ribozyme cleavage activity.

The screening methods of the instant invention can be performed in vitro or in vivo (e.g., in cultured cells) . The glmS ribozymes can be introduced into cells, for example, directly or encoded for in a vector. Alternatively, the native glmS ribozyme may be employed, wherein cleavage products can be identified, for example, by the use of probes specific for glmS. The glmS ribozyme of the instant invention can be the full length glmS ribozyme as depicted in Figure IA. Additionally, the glmS ribozyme may also be a derivative of the glmS ribozyme, such as, without limitation, glmS ribozymes comprising partial or complete deletions of the P4 and/or P3 regions. Nucleic acid molecule (s) may be linked to the glmS ribozyme at either its 3' or 5' ends .

Notably, the glmS gene is conserved among Gram positive bacteria (Winkler et al . (2004) Nature 428:281- 286; Barrick et al . (2004) Proc . Natl. Acad. Sci . U.S.A., 101:6421-6426). The glmS gene encodes glutamine-fructose-6-phosphate aruidotransferase. This enzyme catalyzes a reaction in the pathway of the production of UDP-GIcNAc, a component employed in cell

wall biosynthesis. Accordingly, test compounds identified as modulating glmS ribozyme activity (thereby modulating glmS RNA cleavage and glmS expression) are candidate antibacterial agents . In a particularly preferred embodiment, the antibacterial agents are modulators that increase glmS ribozyme activity. The antibacterial properties of modulators of glmS ribozyme activity can be further tested by incubating the compound with certain bacteria and by other methods known in the art .

In accordance with another aspect of the instant invention, the glmS ribozyme may be employed like an allosteric ribozyme. For example, the glmS ribozyme can be operably linked to a nucleic acid of interest such that the self-cleavage activity of the glmS ribozyme releases the nucleic acid molecule of interest. In a particular embodiment of the instant invention, the self-cleavage of the glmS ribozyme operably linked to a nucleic acid of interest is modulated by a compound other than GlcNβP, such as, for example, serinol or the other coenzymes provided herein.

In a particular embodiment of the invention, the nucleic acid molecule of interest lined to the glmS ribozyme can be part of an siRNA and, thus, be employed to generate siRNA molecules as exemplified in WO 2005/001039.

Other methods of use for the glmS ribozyme ("riboswitch") , such as the regulation of gene expression, are disclosed in U.S. Patent Application 10/569,162. For example, the glmS ribozyme may be operably linked to gene of interest by being placed in the 5'UTR of the gene such that cleavage by the ribozyme prevents expression of the gene.

Also provided herein are ligands/coenzymes which modulate the activity of a glmS ribozyme. In a particular embodiment, the modulating coenzymes comprise the shaded motif of GlcNβP depicted in Figure 2A. Exemplary modulating coenzymes include, without limitation, GIcN, serinol, and GlcNδP analogs. In a particular embodiment, the GlcNβP analogs are modified such that they are more resistant to phosphatase mediated cleavage. Such modifications include, without limitation, phosphonates and halogenated phosphonates of the native phosphate group (see, e.g., Berkowitz et al . (2000) J. Org. Chem. , 65:4498-4508; Berkowitz et al . (1996) J. Org. Chem., 61:4666-4675). Figure 2E provides the structures of the phosphonate analog, fluorinated phosphonate analog, and difluorinated phosphonate analog of GlcNβP.

The present invention also encompasses kits for practicing the methods of the instant invention. Such kits may comprise a recombinant vector containing a nucleic acid sequence encoding a glmS ribozyme operably linked to a promoter suitable for expression in a desired host cell or in vitro transcription. Optionally, the glmS ribozyme may also be operably linked to a multiple cloning site suitable for cloning a nucleic acid molecule of interest so that the cleavage activity of the ribozyme produces the nucleic acid molecule of interest and/or modulates the expression of the nucleic acid molecule of interest (e.g., modulates the expression of the polypeptide encoded by the nucleic acid molecule of interest) . The promoter is preferably a strong promoter and may be constitutive or regulated. Such promoters are well known in the art and include, but are not limited to, RNA polymerase II promoters, the T7 RNA polymerase promoter, and the RNA polymerase III

promoters Uβ and Hl. The kit may also contain more than one recombinant vector wherein the vectors comprise different multiple cloning sites. The recombinant vectors may be provided in any suitable buffer. The kits may further comprise a composition comprising at least one modulator of the glmS ribozyme. In a particular embodiment, at least one modulator is not GlcNβP. The kits may also comprise buffers suitable for the cleavage of the ribozyme-nucleic acid molecule, frozen stocks of host cells, and instruction material. The kit may also further comprise at least one reagent suitable for the in vitro transcription of the recombinant vector. The kits may also further comprise at least one reagent suitable for introducing the recombinant vector or transcribed RNA product into test samples, cells, or subjects.

As used herein, an "instructional material" includes a publication, a recording, a diagram, or any other medium of expression which can be used to communicate the usefulness of the composition of the invention for performing a method of the invention. The instructional material of the kit of the invention can, for example, be affixed to a container which contains a kit of the invention to be shipped together with a container which contains the kit. Alternatively, the instructional material can be shipped separately from the container with the intention that the instructional material and kit be used cooperatively by the recipient.

Definitions

The following definitions are provided to facilitate an understanding of the present invention:

Although exemplified herein as the Bacillus cereus glmS ribozyme, the terms "glmS" , "glmS ribozyme", n glmS

gene" include the gliaS products of other bacteria, particularly other Gram positive bacteria. In a particular embodiment, the glmS ribozyme has 75%, 80%, 85%, 90%, 95%, or 98% identity with SEQ ID NO: 1, particularly nucleotides +1 to +67 as depicted in Figure IA, and has cleavage activity which can be modulated by GlcNβP.

"Nucleic acid" or a "nucleic acid molecule" as used herein refers to any DNA or RNA molecule, either single or double stranded and, if single stranded, the molecule of its complementary sequence in either linear or circular form. In discussing nucleic acid molecules, a sequence or structure of a particular nucleic acid molecule may be described herein according to the normal convention of providing the sequence in the 5 ' to 3' direction. With reference to nucleic acids of the invention, the term "isolated nucleic acid" is sometimes used. This term, when applied to DNA, refers to a DNA molecule that is separated from sequences with which it is immediately contiguous in the naturally occurring genome of the organism in which it originated. For example, an "isolated nucleic acid" may comprise a DNA molecule inserted into a vector, such as a plasmid or virus vector, or integrated into the genomic DNA of a prokaryotic or eukaryotic cell or host organism.

When applied to RNA, the term "isolated nucleic acid" may refer to an RNA molecule encoded by an isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been sufficiently separated from other nucleic acids with which it would be associated in its natural state (i.e., in cells or tissues) . An isolated nucleic acid (either DNA or RNA) may further represent a molecule produced

directly by biological or synthetic means and separated from other components present during its production.

A "vector" is a replicon, such as a plasmid, cosmid, bacmid, phage or virus, to which another genetic sequence or element (either DNA or RNA) may be attached so as to bring about the replication of the attached sequence or element.

The term "probe" as used herein refers to an oligonucleotide, polynucleotide or DNA molecule, whether occurring naturally as in a purified restriction enzyme digest or produced synthetically, which is capable of annealing with or specifically hybridizing to a nucleic acid with sequences complementary to the probe. A probe may be either single-stranded or double-stranded. The exact length of the probe will depend upon many factors, including temperature, source of probe and use of the method. For example, for diagnostic applications, depending on the complexity of the target sequence, the oligonucleotide probe typically contains 15-25 or more nucleotides, although it may contain fewer nucleotides. The probes herein are selected to be complementary to different strands of a particular target nucleic acid sequence. This means that the probes must be sufficiently complementary so as to be able to "specifically hybridize" or anneal with their respective target strands under a set of pre-determined conditions. Therefore, the probe sequence need not reflect the exact complementary sequence of the target. For example, a non-complementary nucleotide fragment may be attached to the 5' or 3 ' end of the probe, with the remainder of the probe sequence being complementary to the target strand. Alternatively, non-complementary bases or longer sequences can be interspersed into the probe, provided that the probe sequence has sufficient complementarity

with the sequence of the target nucleic acid to anneal therewith specifically.

With respect to single stranded nucleic acids, particularly oligonucleotides, the term "specifically hybridizing" refers to the association between two single-stranded nucleotide molecules of sufficiently complementary sequence to permit such hybridization under pre-determined conditions generally used in the art (sometimes termed "substantially complementary"). In particular, the term refers to hybridization of an oligonucleotide with a substantially complementary sequence contained within a single-stranded DNA molecule of the invention, to the substantial exclusion of hybridization of the oligonucleotide with single- stranded nucleic acids of non-complementary sequence. Appropriate conditions enabling specific hybridization of single stranded nucleic acid molecules of varying complementarity are well known in the art.

For instance, one common formula for calculating the stringency conditions required to achieve hybridization between nucleic acid molecules of a specified sequence homology is set forth below (Sambrook et al. , 1989) :

T n , = 81.5 C + 16.6Log [Na+] + 0.41(% G+C) - 0.63 (% formamide) - 600/#bp in duplex

As an illustration of the above formula, using [Na+] = [0.368] and 50% formamide, with GC content of 42% and an average probe size of 200 bases, the T m is 57 0 C. The T m of a DNA duplex decreases by 1 - 1.5 0 C with every 1% decrease in homology. Thus, targets with greater than about 75% sequence identity would be observed using a hybridization temperature of 42 0 C. The stringency of the hybridization and wash depend primarily on the salt concentration and temperature of

the solutions. In general, to maximize the rate of annealing of the probe with its target, the hybridization is usually carried out at salt and temperature conditions that are 20-25 0 C below the calculated T m of the hybrid. Wash conditions should be as stringent as possible for the degree of identity of the probe for the target. In general, wash conditions are selected to be approximately 12-2O 0 C below the T m of the hybrid. In regards to the nucleic acids of the current invention, a moderate stringency hybridization is defined as hybridization in 6X SSC, 5X Denhardt ' s solution, 0.5% SDS and 100 μg/ml denatured salmon sperm DNA at 42 0 C, and washed in 2X SSC and 0.5% SDS at 55 0 C for 15 minutes. A high stringency hybridization is defined as hybridization in 6X SSC, 5X Denhardt ' s solution, 0.5% SDS and 100 μg/ml denatured salmon sperm DNA at 42 0 C, and washed in IX SSC and 0.5% SDS at 65DC for 15 minutes. A very high stringency hybridization is defined as hybridization in 6X SSC, 5X Denhardt ' s solution, 0.5% SDS and 100 μg/ml denatured salmon sperm DNA at 42 0 C, and washed in 0. IX SSC and 0.5% SDS at 65 0 C for 15 minutes .

The phrase "small, interfering RNA (siRNA) " refers to a short (typically less than 30 nucleotides long) double stranded RNA molecule. Typically, the siRNA modulates the expression of a gene to which the siRNA is targeted.

The phrase "operably linked, " as used herein, may refer to a nucleic acid sequence placed into a functional relationship with another nucleic acid sequence. Examples of nucleic acid sequences that may be operably linked include, without limitation, promoters, transcription terminators, enhancers or activators and heterologous genes which when transcribed

and, if appropriate to, translated, will produce a functional product such as a protein, ribozyme or RNA molecule. The phrase "operably linked" may also refer to a nucleic acid sequence placed in functional relationship with a ribozyme such that the catalytic cleavage activity of the ribozyme leads to the release of the operably linked nucleic acid sequence.

The term "antibacterial" refers to having any deleterious effects upon a microbe or bacteria, respectively. Examples include, but are not limited to, inhibition or prevention of growth or reproduction, killing, and inhibiting any metabolic activity of the target organisms.

The following examples describe illustrative methods of practicing the instant invention and are not intended to limit the scope of the invention in any way.

EXAMPLE 1: Templates for transcription were prepared by primer extension and PCR amplification using synthetic DNA corresponding to ribozyme sequence. Ribozymes were prepared by in vitro transcription using T7 RNA polymerase and 32 P-labeled by incorporation of [α- 32 P] - UTP. Transcription products were separated by denaturing 10% polyacrylamide gel electrophoresis (PAGE) and ribozymes were eluted in solution containing 50 mM HEPES (pH 7.5 at 23°C) and 200 mM NaCl, precipitated with ethanol, and redissolved in water. Ribozyme reactions contained ≤250 nM ribozyme and ligand as indicated and were performed under standard conditions consisting of incubation, typically for 2 hours, at 23 0 C in solution containing 50 mM HEPES (pH 7.5 at 23 0 C) and 10 mM MgCl 2 except as noted otherwise.

TRIS, MES, PIPES, TAPS, or CHES was substituted for HEPES as indicated. Reactions were terminated by the addition of a gel loading dye containing 10 M urea and 20 mM EDTA. Products were separated by denaturing 10% PAGE and analyzed using a PhosphorImager and IMAGEQUANT® software (Molecular Dynamics) . Jc obs values for self- cleavage were derived by plotting the natural logarithm of the fraction of uncleaved ribozyme versus time and establishing the negative slope of the resulting line. Stated values represent the average of two replicate assays. Apparent Ki was established by fitting data from three replicate assays to an exponential curve. Apparent K D values were established by fitting average data from at least two replicate asssays to the Michaelis-Menton equation: k ot , s = (k max X

[ligand] ) / ( [ligand] + apparent K n ) . Apparent pK a values were determined by fitting average data from at least two replicate assays to the Hill equation: k O b s = (k max X pH n ) / (pH n + apparent pK a 11 ) , where n represents the Hill coefficient.

To discriminate between the potential roles for GlcNβP in the mechanism of glmS ribozyme self-cleavage, the self-cleavage activity of the glmS ribozyme derived from Bacillus cereus was examined (Fig. IA) . The glmS ribozyme contains four paired domains (P1-P4) that are highly conserved among an 18-member phylogeny (Barrick, et al. (2004) Proc . Natl. Acad. Sci . 101:6421-6426). Previous characterization of the B. subtilis glmS ribozyme demonstrated that the minimal segment required to establish GlcN6P-dependent self-cleavage activity encompasses nucleotides from the +1 position (relative to the cleavage site) through the P2 domain, while the P3-P4 domains serve to enhance activity (Winkler et al . (2004) Nature 428:281-286). For the full-length B.

subtilis glmS ribozyme, GlcNβP binding was reported to elicit a 1000-fold increase in the observed rate constant (k O bs) for self-cleavage activity under saturating GlcNβP and divalent metal ion concentrations (Winkler et al . (2004) Nature 428:281-286). However, this analysis of the B. cereus glmS ribozyme reveals that GlcNβP is more vital to self-cleavage activity than previously appreciated. '

The self-cleavage reaction of the P1-P4 B. cereus glmS ribozyme in the absence or presence of 200 μM GlcNβP for 2 hours (Fig. IB) demonstrates that the ribozyme possesses considerable self-cleavage activity in the absence of ligand in TRIS-buffered solution as previously reported (Winkler et al . (2004) Nature, 428:281-286). To the contrary, the ribozyme is largely devoid of self-cleavage activity in HEPES-buffered solution in the absence of GlcNβP (Fig. IB) , demonstrating a greater dependence upon GlcNβP. Similar analysis of a truncated P1-P3 ribozyme (nucleotides -13 through +86 of the B. cereus glmS ribozyme; Fig. 1C) demonstrates that TRIS-activated self-cleavage is more pronounced for the P1-P4 ribozyme, consistent with the previous observation that the P3-P4 domains function to enhance activity (Winkler et al . (2004) Nature, 428:281- 286). By varying the molar ratio of HEPES: TRIS in otherwise equivalently buffered solutions (Fig. ID) , it is evident that the self-cleavage activity of the P1-P4 ribozyme responds to TRIS concentration in a manner similar to GlcN6P-dependent activation (Winkler et al . (2004) Nature, 428:281-286). These data indicate that TRIS can functionally substitute for GlcNβP, although it appears to be a lower-affinity ligand incapable of saturating the catalyst at concentrations ≤ 50 mM. Determination of k ohs values for ribozyme self-cleavage

in the absence or presence of TRIS (Fig. IE) demonstrates that 10 itiM TRIS accounts for a rate enhancement of 130-fold over the background Jc obs of ~10 ~5 min "1 (Table 1) . By comparison, saturating 10 itiM GlcNβP elicits a k ohs of 1.1 min "1 (Winkler et al . (2004) Nature, 428:281-286), revealing that the true rate enhancement provided by GlcNβP binding to the glmS ribozyme is approximately 5 orders of magnitude. Importantly, the background Jc ObS for the ribozyme approaches that for uncatalyzed transesterification of an unconstrained phosphodiester linkage (Soukup and Breaker (1999) RNA, 5:1308-1325), suggesting that the catalytic mechanism of the glmS ribozyme is intrinsically linked to GlcNβP.

Apparent Rate

Ligand ■fcob s (miiT 1 ) K D (iriM) Enhanc emen t

GlcNβP ~3 -0 . 2 ' ' 300 , 000

GIcN 3 . 2 x 10 "1 > 5 32 , 000

Serinol 7 . 5 x 10- 5 > 5 750

TRIS 1 . 3 x 10 ~J > 50 130

- ~10 ~b - — Table 1. Kinetic parameters for the glmS ribozyme in the absence or presence of 10 itiM GlcNβP or various analogs.

In consideration of a molecular basis for TRIS- dependent glmS ribozyme self-cleavage, it is noteworthy that TRIS structure is analogous to the amine-containing portion of GlcNβP. Such a finding implicates the importance of the amine functionality in ligand binding and activation of self-cleavage. An analysis of GlcNβP and related compounds (Pig. 2A) reveals a striking dependence of ribozyme activity upon amine-containing analogs (Fig. 2B) . Both GlcNβP and GIcN activate glmS ribozyme self-cleavage while GlcβP and Glc do not.

Notably, GIcN is one order of magnitude less potent than GlcNβP as an activator (Table 1) . These data demonstrate that the phosphate moiety of GlcNβP is to some extent expendable with regard to binding and catalysis, while the amine functionality is vital.

Further examination of amine-containing analogs reveals the simplicity of ligand recognition and catalysis (Figs. 2A and 2B) . Serinol, a more precise analog of GlcN than TRIS, expectedly functions as a more potent activator of ribozyme activity (Table 1) while TMG, an analog lacking the amine group, fails to elicit ribozyme self-cleavage. Interestingly, L-ser supports ribozyme activity while D-ser does not. These compounds demonstrate that stereochemical presentation of amine and hydroxyl functionalities is an important determinant of ligand recognition and catalysis, and implicate the importance of the Cl hydroxyl of GlcN6P. Moreover, adjacent amine and hydroxyl functionalities establish the minimal requirement for activity as EtOHN supports ribozyme self-cleavage while EtOH, MeN, and ammonium ion do not. These data clearly demonstrate the glmS ribozyme' s minimal but rigid chemical requirement of ligand to perform catalysis .

Discriminating between the effects of functional group substitutions on ligand binding versus catalysis is a formidable challenge. This is particularly the case for GlcNβP activation of the glmS ribozyme, where there appear to be few determinants for ligand recognition and catalysis which likely attribute to the relatively poor apparent dissociation constant [K 0 ) determined for GlcNβP (Winkler et al . (2004) Nature, 428:281-286). Hydroxyl group substitution of the amine functionality by GlcβP does not entirely preclude the possibility of hydrogen-bonding at the affected position

(i.e. binding). However, the relatively greater acid dissociation constant (pJT a ) of the hydroxyl group markedly reduces the potential of GlcβP to effect proton transfer in a general acid-base mechanism of catalysis. Indeed, GlcβP is observed to function as a competitive inhibitor of GlcNβP analog-dependent glmS ribozyme self- cleavage (Fig. 2C) , therefore demonstrating competence in binding but not catalysis. The extent of competitive inhibition likely reflects the relative binding affinity of each analog versus that of GlcβP, as TRIS-, L-ser-, and EtOHN-dependent self-cleavage are more easily competed with GlcβP than are GIcN- and serinol-dependent self-cleavage. While GlcβP is unable to effect saturating, GlcNβP-dependent self-cleavage (Fig. 2C) , it does effectively inhibit nonsaturating, GlcNβ-dependent self-cleavage with an apparent Ki of 3.0 mM (Fig. 2D) . The absolute inability of the hydroxyl group substitution to partially support catalysis in GlcβP- ribozyme complexes indicates that the function of the amine group in GlcNβP-activated self-cleavage is not attributable to binding alone. These data therefore demonstrate that the amine functionality is more important to catalysis than binding per se, and argue strongly in favor of coenzyme function for GlcNβP in glmS ribozyme self-cleavage .

The hypothesis that coenzyme function of GlcNβP in glmS ribozyme self-cleavage involves proton transfer by the amine functionality in a general acid-base mechanism of catalysis predicts that ribozyme activity is necessarily dependent upon the &K a of the amine group. Advantageously, p.?C a values determined for GIcN and serinol (Fig. 4) differ by ~1 (Fig. 2A) , thereby enabling a facile means of assessing coenzyme function by examining the reactivity profiles of ribozyme with

each ligand. At constant pH, varying the concentration of GIcN or serinol produces a linear reactivity profile with a slope of ~1 (Fig. 3A) , suggesting that ribozyme is not saturated at GIcN or serinol concentrations ≤ 10 iriM, but consistent with previous data for GlcNβP demonstrating that activation results from a single binding event (Winkler et al . (2004) Nature, 428:281- 286) . Moreover, the reactivity profiles for GIcN and serinol exhibit a relative shift by one log unit ligand concentration. This shift is likely attributable to the unit difference in pjf a between the two ligands (i.e. log unit difference in the concentration of deprotonated species at constant pH below pX a ) , and suggests that deprotonated ligand is the effective activator. Accordingly, the reactivity profiles generated by varying pH with constant GIcN or serinol concentration (Fig. 3B) exhibit a slope of ~1 and a relative shift by one pH unit under reaction conditions where pH does not exceed the pJC a of either compound. These data demonstrate that there is a single deprotonation event for GIcN or serinol that influences ribozyme activity and corresponds with the established unit difference in Pi 5 Ca between the two compounds. Importantly, reactivity reaches a near identical maximum under conditions where pH exceeds the pK a of each compound, demonstrating that fully deprotonated GIcN and serinol function equivalently with regard to binding and activation of catalysis. Moreover, the approximate pH required for each ligand to elicit a half-maximal rate constant for self-cleavage (Fig. 3B) corresponds closely to the pJ=C a values of each compound. Reactivity profiles therefore demonstrate a relationship between protonation state of ligand and ribozyme activity that support the role of

ligand as a coenzyme in a general acid-base mechanism of catalysis.

Interestingly, the reactivity profile for glmS ribozyme activation by GlcNβP exhibits a markedly different relationship with pH (Fig. 3B) . The pH- reactivity profile was examined with nonsaturating GlcNβP, where the apparent pKa is 7.8. Importantly, the amine functionality of GlcNβP alone exhibits a pKa of 8.2 (Figure 5). The data, therefore, suggest that the pKa of GlcNβP is perturbed slightly toward neutrality in the context of the ribozyme; a feature expected to enhance catalysis. Considering that the ribozyme appears incapable of similarly affecting the pKa of either GIcN or serinol as ligand analogs suggests that it works specifically in concert with GlcNβP to maximize catalysis. While the phosphate moiety of GlcNβP enhances binding affinity (Table 1) , interaction of the ligand with the ribozyme might "mask" the proximity effect of the phosphate group on the pKa of the amine functionality, thereby yielding an apparent pKa for

GlcNβP that is consistent with the apparent and actual pKa values for GIcN (Figures 4B and 5) . Importantly, the apparent pKa for the GlcNβP stimulated reaction likely accurately reflects the pKa of GlcNβP rather than an effect of the protonation state upon ligand binding affinity, as the apparent K D for GlcNβP is affected only 3-fold by pH over a range that corresponds to a 20-fold difference in the ratio of deprotonated and protonated GlcNβP in solution (Figures 3D and 3F) . Moreover, the apparent pKa of the reaction is not affected by GlcNβP concentration (Figures 3E and 3G) . The data demonstrate that the protonation state of the amine functionality has a marginal effect on ligand binding although it is crucial to catalysis, and therefore support the

possibility that ligand affects catalysis through its capacity for proton transfer rather than binding alone. Consequently, the data are consistent with the hypothesis that GlcNβP functions as a coenzyme rather than an effector in glmS ribozyme self-cleavage.

To slow the reaction kinetics of the glmS ribozyme, the pH-reactivity profile with GlcNβP was performed with no added metal ion other than that which is inherent to the GlcNβP preparation (Fig. 5) . Under the condition of limiting metal ion concentration estimated to be 1 mM, the approximate pH required to elicit a half-maximal rate constant for ribozyme self-cleavage is 7.4. This pH-reactivity profile is similar to that previously determined under conditions of saturating magnesium ion concentration (Winkler et al . (2004) Nature, 428:281- 286) . Importantly, the amine functionality of GlcNβP alone exhibits a Pi 5 T 3 of 8.2 that is unaffected by added magnesium ion (Fig. 4) . The data therefore indicate that the ribozyme is capable of substantially perturbing the p-K a of GlcNβP toward neutrality to enhance catalysis. Figure 4 provides the results for the determination of pK a values for GlcNβP, GIcN, and serinol . Figure 4A provides potentiometric titration curves for 10 mM GIcN (B) and serinol (A) . The x-axis represents equivalents of added 1 N NaOH. The reciprocal of the change in pH per equivalent of added NaOH (1/δpH) is plotted for GIcN (D) and serinol (C) to help identify the midpoint of each curve corresponding to the approximate pi? a value (indicated) . The pJC a determined for GIcN is in precise agreement with that of Park & Choi (Park and Choi (1983) Bulletin Korean Chem. Soc . , 4:68-72) rather than Setnikar et al . (Setnikar et al . (1986) Arzneimittelforschung, 36:729-735) .

Figure 4B provides the potentiometric titration curves for 1 iriM GlcNβP in the absence or presence of 1 iriM MgCl 2 (B and A 7 respectively) . Approximate pfC a values were determined as described above. The lower value is consistent with the pK a of a phosphate ester (Blackburn and Gait (1996) Nucleic acids in chemistry and biology (Oxford University Press, New York), while the higher value represents the pK a of the amine functionality. The relative increase in pX a for the amine functionality of GlcNβP compared to GIcN is consistent with the proximity effect of a negatively charged phosphate moiety (Blackburn and Gait (1996) Nucleic acids in chemistry and biology (Oxford University Press, New York) .

Figure 5 provides reactivity profiles of GlcNδP- activated glmS ribozyme self-cleavage in the absence of added metal ion. Figure 5A demonstrates the GlcN6P concentration-dependence of the reaction. Reactions were performed under standard conditions excluding MgCl2 with various concentration of GlcN6P. Resulting activity is attributable to metal ion inherent to the GlcNδP preparation. Figure 5B demonstrates the EDTA- concentration dependence of the reaction. Reactions were performed under standard conditions excluding MgCl2 with 10 mM GlcN6P and various concentrations of EDTA. Reaction is effectively inhibited by 1 mM EDTA, therefore indicating the inherent metal ion concentration to be approximately 10% that of GlcN6P.

EXAMPLE II

The glmS ribozyme is particularly intriguing among self-cleaving ribozymes with regard to its use of a nonsubstrate ligand to promote catalysis. As shown hereinabove, the ribozyme absolutely requires ligand to

achieve catalysis at rates greater than that of background transesterification. Furthermore, catalysis is dependent upon the presence and the acid dissociation constant (K a ) of the amine group in GlcNβP and related ligand analogs. Therefore, GlcNδP has been proposed to function as a coenzyme in glmS ribozyme self-cleavage that might effect an acid-base catalysis mechanism. However, there are currently no data revealing the nature or location of the ligand-binding site with respect to the catalytic core of the ribozyme.

To investigate metal ion binding and ligand interaction in the B. cereus glmS ribozyme, nucleoside analog interference mapping (NAIM) and nucleoside analog interference suppression (NAIS) were performed (Ryder, and Strobel (1991) Methods 18:38-50; Ryder et al . (2000) Methods Enzymol . 317:92-109). Backbone and nucleobase analogs were used to identify functional groups required for GlcNβP-dependent ribozyme activity. Essential phosphate oxygens, 2'-hydroxyls and purine nucleobase groups indicate sites of metal ion or ligand interaction, or tertiary contacts within the ribozyme. Thiophilic metal ion rescue of phosphorothioate interference was used to distinguish metal ion-binding sites (Waring, R. B. (1989) Nucleic Acids Res., 17:10281- 10293; Ruffner and Uhlenbeck (1990) Nucieic

Acids Rer. 18:6025-6029 1990; Baru and Strobel (1999) RNA 5:1399-1407). Furthermore, NAIS was used to distinguish contacts between ribozyme functional groups and the phosphate moiety of the ligand, on the basis of interferences that are suppressed when glmS ribozyme self-cleavage is stimulated by glucosamine (GIcN) . These studies demonstrate that there are a number of requisite metal ion contacts within the catalytic core, some of which are affected by peripheral domains of the

ribozyme. Moreover, backbone and nucleobase functional groups within the catalytic core that mediate ligand recognition lie close to the cleavage site, consistent with the hypothesis that ligand functions in the active site as a coenzyme to promote self-cleavage.

METHODS

Templates for transcription were prepared by primer extension and PCR amplification using synthetic DNA corresponding to either the P1-P3 (through nucleotide +86) or P1-P4 B. cereus glmS ribozymes (Fig. IA) . Ribozymes were prepared by in vitro transcription using T7 RNA polymerase and 32 P-labeled by incorporation of [α- 32 P]UTP. Transcription products were separated by denaturing 10% PAGE and visualized by UV shadowing.

Ribozymes were excised and eluted in solution containing 50 mM HEPES (pH 7.5 at 23 0 C) , I mM EDTA and 200 mM NaCl, precipitated with ethanol, redissolved in water and quantified by scintillation counting. Riboyme reactions were incubated at 23 0 C and contained -0.2 μM RNA, 50 mM HEPES (pH 7.5 at 23 0 C), 200 μM GlcNβP and varying concentrations of MgCl2. Aliquots from reactions were terminated at various time points by the addition of an equal volume of gel loading dye containing 10 M urea and 20 mM EDTA. Products were separated by denaturing 10% PAGE and analyzed using a PhosphorImager and IMAGEQUANT® software (Molecular Dynamics) . Observed rate constants (k Obs values) of self-cleavage were determined by plotting the natural logarithm of the fraction of uncleaved ribozyme versus time and establishing the negative slope of the resulting line.

Ribozymes containing phosphorothioate nucleotide analogs were prepared from DNA templates described above

by in vitro transcription using T7 KNA polymerase in solution containing 50 mM HEPES (pH 7.5 at 23 0 C), 15 iriM MgCl 2 , 2 mM spermidine, 5 mM DTT, 1 mM each NTP and phosphorothioate nucleotide analog at a concentration previously determined to yield -5% incorporation (Ryder and Strobel (1991) Methods 18:38-50; Ryder et al . (2001) Methods Enzymol., 317:92-109). Ribozymes containing 2'- deoxy phosphorothioate nucleotide analogs were prepared from DNA templates described above by in vitro transcription using Y639F mutant T7 RNA polymerase in solution containing 40 mM HEPES (pH 7.5 at 23 0 C), 15 mM MgCl 2 , 4 mM spermidine, 10 mM DTT, 0.05% (v/v) Triton-X, 1 mM each NTP and phosphorothioate nucleotide analog at a concentration previously determined to yield ~5% incorporation (Ryder and Strobel (1991) Methods 18:38- 50; Ryder et al . (2001) Methods Enzymol., 317:92-109). Ribozymes were purified by denaturing PAGE as described above .

Ribozyme reactions for NAIM were incubated for 45 seconds at 23 0 C and contained ~2 μM analog-containing RNA, 50 mM Tris-HCl (pH 7.5 at 23°C) , 200 μM GlcNβP and 5 mM MgCl 2 . Reactions represented selective conditions that provided ligand at a concentration near the apparent K d and allowed -50% ribozyme self-cleavage . Manganese-rescue experiments were performed by substituting 5 mM manganese acetate for MgCl 2 , and no change in reaction time was required to achieve 50% ribozyme self-cleavage .

Inactive and active RNAs were purified by denaturing PAGE as described above. After treatment of inactive RNAs with alkaline phosphatase, inactive and active RNAs were 5 ' end-labeled with T4 polynucleotide kinase and [γ- 32 P]ATP. Labeled RNAs were again purified by denaturing PAGE and quantified by scintillation

counting. Sites of phosphorothioate analog incorporation were determined by cleavage with 10 mM iodoethane and separation of products by denaturing PAGE. Resulting patterns of analog incorporation for inactive and active RNAs were analyzed using a

Phosphorimager and IMAGEQUANT® software (Molecular Dynamics) . Analog interferences, calculated as K values, were determined as described (Ryder and Strobel (1991) Methods 18:38-50; Ryder et al . (2001) Methods Enzymol . , 317:92-109) and represent averages of at least two replicate assays.

NAIS reactions were incubated for 35 minutes at 23 0 C and contained ~2 μM analog-containing RNA, 50 mM HEPES (pH 7.5 at 23°C) , 500 μM GIcN and 5 mM MgCl 2 . Reactions represented selective conditions that provided ligand at a previously determined nonsaturating concentration and allowed ~50% ribozyme self-cleavage . One exception was that inosine-containing ribozymes required reaction for 90 minutes to achieve 50% self- cleavage. Inactive and active RNAs were then purified, labeled and analyzed as described above.

RESULTS

Divalent metal ions are used by many ribozymes to promote folding and catalysis. Recent studies indicate that the P3 and P4 peripheral domains affect glmS metal ion dependence, as truncated glmS ribozymes show greater magnesium ion dependence than does full-length ribozyme. Therefore, the metal ion dependence of B. cereus glmS ribozymes containing or lacking the P4 domain was examined. The rate constant for ligand-dependent self- cleavage of the P1-P3 ribozyme increases nearly three orders of magnitude as the Mg 2+ concentration rises from 1 mM to 50 mM, whereas the activity of the P1-P4

ribozyme changes less than an order of magnitude across the same Mg 2+ concentration range (Fig. 6) . Therefore, it seems that the P4 domain alters either metal ion affinity or the number of requisite metal ion contacts. Notably, previous studies of glmS ribozyme self-cleavage have shown that Mg 2+ can be substituted by other divalent ions, including Mn 2+ (Winkler et al . (2004) Nature 428:281-286). On the basis of these observations, it can be reasoned that differential metal ion utilization by P1-P3 and P1-P4 ribozymes could be examined by phosphorothioate-analog interference and thiophilic metal ion rescue to elucidate the importance of metal ion interactions within the glmS ribozyme.

NAIM using phosphorothioate analogs enables the identification of possible metal ion-binding sites, as substitution of sulfur for one of the nonbridging phosphate oxygens can interfere with important metal ion contacts that are best supplied by magnesium. Moreover, NAIM can then serve to strengthen evidence for metal ion-binding sites through rescue of phosphorothioate interferences by substitution of the relatively thiophilic divalent metal ion Mn 2+ for Mg 2+ . Phosphorothioate interferences determined for the P1-P3 and P1-P4 ribozymes (Fig. 7A, 7B) cluster within the catalytic core defined by the Pl and P2 domains (Fig.

8A) . In both P1-P3 and P1-P4 ribozymes, similar but not identical patterns of phosphorothioate effects occur within Pl.1 and the segment joining Pl and P2 (Jl/2) . Strong interference (calculated as a large K value) was observed for position C2 , near the cleavage site of the glmS ribozyme, independent of the presence of the P4 domain (Fig. 8A) . However, for the Jl/2 region, strong phosphorothioate effects were observed for C29, G30 and A31 for the P1-P3 ribozyme, whereas weak effects occur

at A28, G30 and A31 for the P1-P4 ribozyme (Figs. 7 and 8) . Additionally, the P1-P3 ribozyme shows weak phosphorothioate effects in Pl .1 at G3 and G7 , whereas the P1-P4 ribozyme exhibits strong interference at G3. Notably, phosphorothioate effects observed at C52 and G53 in the P2a stem of the P1-P3 ribozyme are entirely absent in the P1-P4 ribozyme (Figs. 7 and 8). These data demonstrate that the P4 domain substantially influences the relative importance of interactions that contribute to the stability of the catalytic core of the glmS ribozyme and suggest that the P4 domain might interact directly with the P2a region to enhance stability and catalysis.

Next, NAIM analyses were performed after substituting Mn 2+ for Mg 2+ ions, to distinguish interferences that represent putative metal ion contacts. For the P1-P3 ribozyme, partial or full rescue of phosphorothioate effects at G7 , G30 and A31 was observed, whereas interferences at C2 , G3 , C29, C52 and G53 persisted (Figs. 7 and 8 and Table 2). In contrast, the P1-P4 ribozyme showed partial rescue of the strong phosphorothioate effects at C2 and G3 (Table 2) and complete rescue of the weak phosphorothioate effects at A28, G30 and A31 (Figs. 7 and 8 and Table 2). These data indicate that most phosphorothioate interferences observed in the Pl .1 and Jl/2 regions probably represent metal ion contacts, although phosphorothioate interference patterns suggest that metal ion utilization might modestly differ between the P1-P3 and P1-P4 ribozymes . For the P1-P3 ribozyme, no manganese rescue was observed when NAIM was performed with 4 mM Mg 2+ and 1 mM Mn 2+ combined. Furthermore, addition of thiophilic monovalent thallium ion resulted in slight rescue only at C29 and had no effect upon

other sites of interference (Baru and Strobel (2001) Methods 23:264-275). These data indicate that metal ion- binding sites in the glmS ribozyme preferentially interact with divalent cations. Moreover, the location of metal ion contacts strictly within the Pl .1 and Jl/2 segments suggests that the tertiary structure required to establish the ligand-binding and catalytic activities of the glmS ribozyme might arise through metal ion- mediated interaction of these segments .

Manganese

Nucleotide Interference Suppression rescue analog Position K value 3 K value 3 K value 3

AaS A28 2.2 1.1:(: L p* A31 2.1 2.2 1.2* :I!

CaS C2 >6.0 >6.0 4.8 s

GaS G3 >6.0 3.2-j- 3.3»

G30 2.6 1.2* dAαS A58 6.0 6.0 6.0 dCαS C29 5.3 6.0 1.8** dGαS G27 2.3 2.0 j 2** G30 6.0 3.0t 1.0** dUαS U59 >6.0 6.0 6.0

7dAσ.S A28 2.5 1.0* nd b

DAPaS A31 3.0 1.6* nd

G57 3.8 nd nd

Table 2 - Summary of interference, interference suppression, and manganese rescue data for the P1-P4 B. cereus glmS ribozyme. a Interference and suppression K values are given in the range from 1.0 to 6.0, where those values much greater than 6.0 are indicated as >6.0. Partial or full suppression is denoted by a single or double dagger, respectively. Partial or full manganese rescue is denoted by a single or double asterisk, respectively. b Not determined.

To identify contacts between the glmS ribozyme and ligand, NAIS was performed using the P1-P4 ribozyme.

NAIS has been successfully used to identify tertiary interactions within other ribozymes (Strobel et al . (1998) Nat. Struct. Biol., 5:60-66; Szewczak et al . (1998) Nat. Struct. Biol., 5:1037-1042; Soukup et al . (2002) Biochemistry 41:10426-10438), where analysis was based on the principle that if a tertiary interaction is disrupted by site-specific deletion or alteration of one functional group in an interacting pair, then no additional energetic penalty will result from deletion or alteration of the second functional group. The typical NAIS methodology has been modified by combining the use of a ligand analog (serving as a site-specific modification) with interference mapping to identify sites at which nucleotide-analog incorporation that previously caused interference with the cognate ligand is tolerated with the ligand analog. NAIS experiments were performed using the ligand analog GIcN, which lacks the 5 ' -phosphate group of GlcN6P, but has previously been shown to stimulate P1-P4 glmS ribozyme activity at a reduced rate. Thus, GlcN-dependent suppression of phosphorothioate interference in the glmS ribozyme reveals contacts between the phosphate moiety of GlcNδP and phosphate oxygens in the RNA catalyst.

NAIS performed in the presence of Mg 2+ using the Pl- P4 glmS ribozyme identified sites of phosphorothioate- interference suppression within the Pl .1 and Jl/2 segments. In Pl.1, GIcN-dependent suppression of phosphorothioate interference was observed at G3 but not at C2 (Fig. 9A, 9B) . For the Jl/2 segment, suppression of phosphorothioate interference was observed at A28 and G30 but not at A31 (Fig. 9C, 9D) . Therefore, NAIS data suggest that three glmS ribozyme phosphate oxygens compose a portion of the ligand-binding site responsible for recognition of the phosphate moiety of GlcN6P.

Moreover, the composition of the ligand-binding site indicates that GlcNβP binds adjacent to the cleavage site, in a catalytic core created by interactions among the highly conserved sequence segments within Pl .1 and Jl/2. Considering that sites of interference suppression also showed metal ion suppression in earlier experiments, ligand phosphate recognition by ribozyme phosphate oxygens is probably mediated by metal ions. These data demonstrate that the ligand-binding and active sites of the glmS ribozyme are inseparable entities and support the view that GlcNβP functions at the active site to promote self-cleavage.

To further investigate the importance of backbone functional groups within the glmS ribozyme, both NAIM and NAIS analyses were performed using 2 ' - deoxyphosphorothioate analogs. Roles for 2 ' -hydroxy1 groups in RNA structure and function have been demonstrated by numerous similar studies of ribozymes (Strobel et al . (1998) Nat. Struct. Biol. 5:60-66; Szewczak et al . (1998) Nat. Struct. Biol., 5:1037-1042). 2 ' -hydroxy1 groups can serve as both hydrogen bond donors and acceptors, and thus can participate in tertiary interactions with nucleobase functional groups or other ribose units to promote folding and catalysis. For the glmS ribozyme, it is therefore possible that 2'- hydroxyl groups might also mediate ligand recognition. NAIM performed in the presence of Mg 2+ and GlcN6P for the P1-P3 and P1-P4 ribozymes identified important 2 ' -hydroxy1 groups within the Pl .1 and Jl/2 regions. Identical 2 ' -deoxy effects were observed within Pl .1 for both constructs, with strong interferences at A58 and U59 (Pig. 8B). In Jl/2, by contrast, strong 2 ' -deoxy effects were observed at G27 and A28 for the P1-P3 ribozyme but at C29 and G30 for the P1-P4 ribozyme (Fig.

4b). Additionally, the P1-P4 ribozyme shows weak 2'- deoxy interference at G27. These data further demonstrate that the P4 domain affects the stability of the catalytic core. However, the possibility that 2'- deoxy effects might occur at C29 and G30 in the P1-P3 ribozyme, as they do in the P1-P4 ribozyme, cannot be excluded, owing to phosphorothioate interferences at those positions. Furthermore, NAIS performed in the presence of Mg 2+ and GIcN for the P1-P4 ribozyme revealed one site of 2 ' -deoxy interference suppression at G30 (Table 2). These data establish an additional contact with ligand phosphate that further defines the GlcNβP- binding site within the catalytic core of the glmS ribozyme . Notably, NAIM performed in the presence of Mn 2+ and GlcNβP for the P1-P4 ribozyme did not show any of the 2 ' -deoxy interferences in Jl/2 that were observed in the presence of Mg 2+ , whereas those in Pl .1 persisted (Fig. 8B and Table 2) . It is conceivable that the relative importance of 2 ' -deoxy substitutions is obscured by additional Mn 2+ interactions with accompanying phosphorothioate substitutions. Nevertheless, the data further demonstrate that the Jl/2 region of the glmS ribozyme is relatively adaptable to environmental factors. In combination with P4 domain-dependent differences in nucleotide-analog interference patterns, these results indicate that the Jl/2 segment is relatively plastic, despite being essential to ligand binding and catalysis. To identify nucleobase groups that are important to ribozyme structure, catalysis and ligand binding, both NAIM and NAIS analyses of the P1-P4 glmS ribozyme were performed in the presence of Mg 2+ using available purine

nucleotide analogs. NAIM using adenosine analogs, including N-methyladenosine (m 6 AαS) , 7-deaza-adenosine (7dAαS) , diaminopurine riboside (DAPaS) and purine riboside (PurαS) , or guanosine analogs, including inosine (IaS) , 7-deaza-guanine (7dGαS) and N- methylguanosine (m 2 GαS) , revealed sites of interference only within the Pl .1 and Jl/2 segments (Fig. 1OA, 10B). Interference was observed at A31 for m 6 AαS, DAPaS and PurαS; A28 for 7dAαS; A58 for DAPaS; G30 far IaS, 7dGαS and m 2 GαS; and G57 for IaS and m 2 GαS . These interference patterns are indicative of contacts to the exocyclic amine of A31, G30 and G57, and to the N7 imine of A28 and G30, whereas steric restraints are implied for A31 and A58 by PurαS interferences. NAIS using GIcN as the ligand analog resulted in either partial or full suppression of each adenosine-analog interference, whereas no suppression was observed for IaS or 7dGαS interferences. Notably, these data, in combination with NAIM and NAIS results from 2 ' -deoxy and phosphorothioate substitutions, demonstrate that nucleotide positions important for ligand binding and catalysis lie exclusively within the Pl .1 and Jl/2 segments of the Pl- P4 glmS ribozyme. This is consistent with the hypothesis that Pl .1 and Jl/2 constitute the catalytic core. Moreover, interference suppression indicative of ligand phosphate interactions is observed at nucleotide positions within each of the three segments that constitute the catalytic care, demonstrating that the 5' Pl.1, 3' Pl .1 and Jl/2 segments indeed converge in a tertiary structure that forms the GlcNβP-binding site near the catalytic cleavage site.

The glmS ribozyme, like many other self-cleaving RNA catalysts, requires metal ions to achieve activity Recent studies suggest that divalent metal ions fulfill

only structural requirements for glmS ribozyme activity, as Mg 2+ can be substituted with a variety of other divalent metal ions or molar concentrations of monovalent metal ion. Moreover, no thiophilic metal ion rescue was observed for a B. cereus glmS ribozyme construct having a phosphorothioate substitution of the cleavage-site phosphate, suggesting that metal ions do not directly effect a mechanism of glmS ribozyme catalysis. Here, five other sites within the catalytic core of the P1-P4 B. cereus glmS ribozyme showed phosphorothioate effects that were rescued by substitution of Mn 2+ for Mg 2+ , where three of the sites also seem to mediate ligand phosphate recognition. Additionally, five sites of 2'-deoxy interference were rescued by substitution of Mn 2+ for Mg 2+ , where one site is involved in ligand phosphate recognition. Manganese rescue experiments suggest two roles for divalent metal ions that are either specific or nonspecific. On one hand, there seem to be specific metal ion interactions with the phosphodiester backbone required to stabilize the active structure. On the other hand, there seem to be nonspecific metal ion effects, particularly in Jl/2, that arise from nucleotide phosphorothioate analog incorporation and that compensate for other functional- group substitutions. Together, these studies demonstrate that divalent metal ions have substantial and specific roles in organization of the catalytic core and ligand recognition, if not catalysis.

Comparison of P1-P3 and P1-P4 B. cereus glmS ribozymes further reveals the intricacies of metal ion utilization. In the P1-P3 ribozyme, phosphorothioate effects are observed in P2a that are entirely absent in the P1-P4 ribozyme. Although these interferences are not rescued by Mn 2+ , the phosphates could represent metal

ion-binding sites/ such as those with highly coordinated magnesium ions, that are not easily rescued by the addition of thiophilic metal ion (Baru and Strobel (1999) KNA 5:1399-1407; Brautigam and Steitz (1998) J. MoI. Biol., 277:363-377). Notably, the reduced metal ion dependence of the P4-containing ribozyme and alleviation of P2a phosphorothioate interferences is consistent with the possibility that P4 aids physiological Mg 2+ binding, and it suggests interaction between the two domains. Furthermore, it's demonstrated that the absence or presence of the P4 domain and the identity of the divalent metal ion influence the pattern of backbone functional groups that are important to ligand binding and catalysis. Dependent on the presence of the P4 domain, there are local shifts in both phosphorothioate and 2 ' -deoxy interferences that are most prevalent in Jl/2. These results suggest that the Jl/2 region shows considerable adaptability in providing for ligand binding and catalysis. This may reflect a complex and extensive network of interactions within the catalytic core, and conditions dictate the relative importance of those interactions required for ligand binding and catalysis.

The combination of NAIM and NAIS applied toward ligand and analog interactions with RNA represents a powerful biochemical means of dissecting molecular recognition. An interesting consideration for the glmS ribozyme is the way anionic ligand phosphate recognition is achieved by a polyanionic macromolecule . It has been demonstrated here that certain sites of phosphorothioate interference that show manganese rescue in NAIM experiments are also identified as sites of ligand phosphate interaction by NAIS. The data strongly suggest that ligand phosphate recognition is mediated by

metal ion coordination with ribozyme phosphate groups. This mode of phosphate recognition is a facile means of interaction that might apply to other riboswitches that bind phosphate-containing ligands such as thiamine pyrophosphate, flavin mononucleotide or adenosylcobalamin. Moreover, metal ion-mediated recognition of GlcNβP by the glmS ribozyme suggests that

I the compound optimally promotes self-cleavage as an organometallic ligand, thus illuminating one role for divalent metal ions in glmS ribozyme structure and function.

NAIM and NAIS analyses further identify 2 ' -hydroxl and nucleobase functional groups that are essential to ribozyme structure, catalysis and ligand recognition. Notably, NAIS data demonstrate that nucleotides within each of the three sequence segments that comprise Pl .1 and Jl/2 are implicated in ligand phosphate recognition as contributing either direct or indirect (that is, metal ion- or water-mediated) contacts, or steric boundaries to the ligand phosphate-binding pocket (Fig. 11) . It is noteworthy that nearly half of the essential functional groups identified by NAIM are indicated by NAIS to contribute to ligand phosphate recognition, indicating a substantial commitment of requisite functional groups toward GlcNβP binding and phosphate recognition. The phosphate moiety of GlcNβP is in part an affinity determinant for the ribozyme, as GIcN supports catalysis but has an indeterminably weaker binding affinity. At present, it is unclear whether the reduced rate of ribozyme self-cleavage afforded by GIcN reflects solely the substantially weaker binding affinity of the ligand analog or includes some contribution of the phosphate moiety that is lacking. Notably, glmS ribozyme self-cleavage has been shown to

depend upon the presence and the K a of the ligand amine group, which is effected by the phosphate moiety. GlcNβP binding has been localized adjacent to the cleavage site and thus highlights the possibility that interplay between ligand phosphate and amine groups directly effects a mechanism of glmS ribozyme self- cleavage, consistent with its proposed role as a coenzyme .

The analysis of essential functional groups within the B. cereus glmS ribozyme using both NAIM and NAIS provides substantial insight regarding the nature of the catalytic core, ligand-binding site and mechanism of GlcNβP-dependent self-cleavage. NAIM data demonstrate that requisite phosphate, 2 ' -hydroxy1 and purine nucleobase groups reside exclusively within the Pl .1 and Jl/2 regions of the P1-P4 ribozyme, suggesting that these highly conserved sequence segments constitute the catalytic core. Moreover, NAIS data show that functional groups in proximity to the phosphate moiety of GlcNβP reside within Jl/2 and Pl.1, revealing that the ligand binds within the catalytic core and adjacent to the cleavage site of the P1-P4 ribozyme. Notably, these results provide evidence that Pl .1 and Jl/2 converge in a tertiary structure that is organized by the putative pseudoknot resulting from Pl .1 interaction. Furthermore, convergence of the Pl .1 and Jl/2 regions to form inseparable ligand-binding and catalytic sites supports the hypothesis that GlcNβP functions as a coenzyme within the glmS ribozyme active site to promote self-cleavage and riboswitch regulation of gene expression.

A number of publications and patent documents are cited throughout the foregoing specification in order to describe the state of the art to which this invention pertains. The entire disclosure of each of these citations is incorporated by reference herein.

While certain of the preferred embodiments of the present invention have been described and specifically exemplified above, it is not intended that the invention be limited to such embodiments. Various modifications may be made thereto without departing from the scope and spirit of the present invention, as set forth in the following claims .