RNA catalysis is of fundamental importance in many biological functions, such as the peptidyl transferase activity of the ribosome and genetic control by riboswitches, among others. Small ribozymes are a convenient system to increase our understanding about the structure, folding and catalytic mechanism of ribozymes. This dissertation includes analysis of certain aspects of the catalytic mechanism in the hairpin and hammerhead ribozyme. In the hairpin ribozyme, we studied the functional consequences of molecular substitutions at two conserved positions, A9 and A10. These nucleotides are located close to the scissile phosphate but their exact function is unclear since they do not appear to be making any essential interactions with other nucleotides in the catalytic core. G, C, U, 2-aminopurine, 2, 6-diaminopurine, purine, and inosine were substituted at A9 and A10 and their effects on cleavage and ligation rates were analyzed. The effect of the variations on tertiary structure and docking was monitored by hydroxyl radical footprinting and native gel electrophoresis. It was observed that all the variants that exhibited poor docking and/or tertiary structure formation were also ligation challenged whereas they performed normally in the cleavage reaction. We found a unique variant, A10G that cleaved five times faster than A10 but did not exhibit any ligation. Results suggested that ligation required a more kinetically stable core than that needed to carry out cleavage. The hammerhead ribozyme field featured extensive disagreements between the crystal structure of the minimal hammerhead released in the mid 90s and the accumulating biochemical data. Much of the conflict was resolved with the new crystal structure of the extended hammerhead ribozyme. This structure confirmed many of the biochemical findings and brought out some new interactions, notably the G8·C3 base pair. We studied numerous base substitutions to establish the importance of the base pair for cleavage and ligation. Catalysis requires the formation of the base pair but even the fastest base paired variant was 10-fold slower than G8·C3 base pair. Docking and tertiary structure analysis by hydroxyl radical footprinting and native gel electrophoresis emphasized the importance of having a purine at position 8 and a pyrimidine at 3. Catalysis in the unmodified ribozyme was uniquely accompanied by hydrolysis of the 2′, 3′- cyclic phosphate ring present on one of the cleavage products, leading to the generation of non-ligatable products during a ligation assay. We determined the ligation rate-pH profile for unmodified ribozyme that differed from the cleavage rate-pH profile only at high pH.

The Varkud Satellite (VS) ribozyme is the largest of the “small” nucleolytic ribozymes and is the only one for which there are no high resolution crystal structures available. The VS ribozyme comprises a catalytic domain and a substrate domain. The catalytic domain includes five helices that interact with the stem-loop substrate. The substrate is docked within a cleft that is formed by helices II and VI. This naturally brings the cleavage site in close proximity to the A730 loop in helix VI. The adenines within the A730 loop are very crucial to the cleavage reaction and any substitution causes a major decrease in the cleavage activity of the ribozyme. This study is aimed at designing and producing a variant of the Varkud Satellite ribozyme that consists of multiple parts that can be used for detailed studies of ribozyme kinetics and assembly.

The Varkud Satellite (VS) ribozyme is the largest of the “small” nucleolytic ribozymes and is the only one for which there are no high resolution crystal structures available. The VS ribozyme comprises a catalytic domain and a substrate domain. The catalytic domain includes five helices that interact with the stem-loop substrate. The substrate is docked within a cleft that is formed by helices II and VI. This naturally brings the cleavage site in close proximity to the A730 loop in helix VI. The adenines within the A730 loop are very crucial to the cleavage reaction and any substitution causes a major decrease in the cleavage activity of the ribozyme. This study is aimed at designing and producing a variant of the Varkud Satellite ribozyme that consists of multiple parts that can be used for detailed studies of ribozyme kinetics and assembly.

The hairpin ribozyme is a small catalytic RNA that cleaves a phosphodiester bond. In order for cleavage to occur, the hairpin ribozyme must properly fold into its docked conformation, in which the two loops interact to form the active site. Metal ions and the four-way junction play critical roles in the stabilization of the docked conformation. The work presented in this thesis attempts to investigate the metal-ion dependence of the docking of the four-way junction construct of the hairpin ribozyme. In addition, the activity of the hairpin ribozyme in the presence of Mn2+ was observed. Initially, a four-stranded four-way junction construct of the hairpin ribozyme and a loopless mutant were characterized by native gel electrophoresis and thermal denaturation to verify ribozyme formation. A novel interaction between the sulfur of a phosphorothioate-substituted mononucleotide, such as adenosine thiomonophosphate (AMPS) or adenosine thiotrisphoshate (ATPgS), and Cd2+ has been characterized by UV-vis spectroscopy. A feature at 208 nm was identified to be a result of sulfur-to-Cd2+ transfer. The apparent binding affinities, the apparent extinction coefficients, and the binding ratios were determined for each complex.

The factors that determine RNA structure formation, stability, and dynamics are inexorably linked to RNA function. The Hammerhead ribozyme (HHRz) has long served as a model for studying metal-dependent folding and catalysis in RNA. The HHRz consists of three helices meeting at a common junction of conserved nucleotides that form the active site of the ribozyme. Current models of metal-dependent HHRz function involve a requirement for divalent metals to globally fold the ribozyme at low metal concentrations, followed by a second metal-dependent process which activates the HHRz for catalysis. The exact role of metal ions in activating HHRz catalysis is still a subject of investigation. We used 2-aminopurine substitutions near the active site of the ribozyme to determine if this second metal-dependent process involves a conformational rearrangement in the core of the ribozyme. We find evidence for a conformational change beyond global folding in the core of the ribozyme that not only correlates with metal activated catalysis but is also sensitive to the identity of the metal ions used for folding. Though phosphorothioate substitutions indicate that a ground-state coordination of a catalytic metal to the scissile phosphate is required for efficient catalysis, our folding studies show that this coordination event is not absolutely required for folding of the HHRz core. To investigate possible roles for metal ions in general acid-base catalysis, we tested the pH dependence of the HHRz rate using a variety of metal ions. We find the pH dependent rate profile of the ribozyme is shifted by transition metal ions, whereas other group II metals show similar profiles to Mg

Metal-RNA interactions are important to neutralize the negative charge and aid in correctly folding the RNA. Spectroscopically active metal ions, especially Mn2+, have been used to probe the type of interaction the metal has with RNA. In previous studies, the hammerhead ribozyme, an RNA motif that catalyzes a site-specific phosphodiester bond cleavage reaction, was determined by room temperature EPR (electron paramagnetic resonance) studies to have a set of tightly and weakly bound metal ions. Under high salt concentrations, the hammerhead was found to bind a single Mn2+ ion with high affinity and with a characteristic low temperature EPR signal. Using site specific 15N labeling of a guanine residue in conjunction with ESEEM (electron spin echo envelope modulation) spectroscopy, the high affinity Mn2+ ion was conclusively determined coordinated to G10.1 of the proposed A9/G10.1 site with four water molecules coordinated to the Mn2+ ion. EPR power saturation studies determined that under low salt conditions the hammerhead coordinates up to four Mn2+ ions in relatively close proximity compared to an RNA duplex. EXAFS (extended X-ray absorption fine structure) spectroscopy was used to determine that a Cd2+ ion coordinates to both the Rp and Sp sulfur atoms of a phosphorothioate modification at the A9 phosphate of the hammerhead. Previous EXAFS results for the Mn2+ substituted A9 phosphorothioate suggested that the Mn2+ ion coordinates to the oxygen atom for both isomers. Molecular modeling suggested that the A9/G10.1 metal site will twist the phosphate group in order to accommodate this coordination. A Mn-GMP and Mn-phosphate model complexes were prepared and characterized by EXAFS to assign the origin of the features observed for the hammerhead sample. A series of RNA sequences with internal loops containing the sheared G-A metal ion binding motif showed greater thermal stabilization of the RNA structure in the presence of Mn2+ ions compared to sequences without the motif. The EPR binding isotherms also showed a set of moderately tight metal ion interaction while circular dichroism spectroscopy was used to investigate structural differences between the sequences. These results suggest a mostly electrostatic, not structural role, for the Mn2+ ion interactions with these sequences.

Group II introns are highly structured RNAs that carry out self-splicing reactions. The multiple turnover version of one of these introns, termed the D135 ribozyme, is derived from the mitochondrial aI5γ intron of Saccharomyces cerevisiae and is widely studied as a model RNA for group II intron folding. An important current goal is to probe global changes during its folding with or without DEAD-box chaperone proteins. My initial experiments to study global compaction using small angle X-ray scattering (SAXS) of D135 reveal rapid initial compaction. Unexpectedly, slower increases in Rg value and forward scattering were observed and shown to result from dimerization of the ribozyme. Dimerization was also observed with native electrophoretic mobility shift assays. Here, I have characterized the dimerization process at various conditions. Dimerization requires Mg2+, with similar concentration dependence as tertiary folding, and the dimer is efficiently disrupted by the ATP-dependent activity of DEAD-box proteins. Dimerization does not affect ribozyme catalysis, as both the monomer and the dimer are shown to be fully active. Further experiments showed that dimerization results from duplex formation by an artificial 3’ tail that has extensive self-complementarity, as the deletion of this tail ablates dimerization. Constructs lacking this artificial 3’ tail are likely to simplify further study of the folding process of this ribozyme. / text

Many different experimental approaches were attempted to achieve carbon-carbon bond formation by nucleic acid template-directed reactions and ribozyme catalysis as potential lipid synthesizing machineries in the RNA world. A novel biomimetic condition for decarboxylative Claisen condensation in polyketide biosynthesis was discovered. The reaction of a malonic acid half oxyester with a Nhydroxysuccinmidyl ester forming reagent resulted in self-condensation to provide the corresponding 1,3-acetonedicarboxylic acid diester in the absence of a divalent metal chelator or a coordinating solvent. The decarboxylative Claisen condensation of malonyl adenosine using a poly-U template in solution or with immobilized poly-U was attempted. Various analytical methods demonstrated that malonyl adenosine underwent an exclusive hydrolysis reaction instead of condensation in the given conditions. Similar results were observed for the reaction of malonyl-CoA with acetyl-CoA on poly-U templates. No evidence for the decarboxylative Claisen condensation was observed by a DNA-templated system although a double helical structure of DNA duplex was proven to facilitate a bimolecular reaction by offering a favorable proximity effect. Therefore, it seems that the unsuccessful condensation resulted not from the bad template effect but from the intrinsic properties of the decarboxylative Claisen condensation reaction itself. Two tRNA molecules loaded with a malonic acid were prepared by ligation of truncated tRNAs with malonylated dinucletides. Our initial attempts to probe carbon-carbon bond formation by subjecting malonylated tRNAs to the in vitro translational machinery were not successful. Novel carbon isosteres of α-amino acids are suggested as a potential source of a more stable and reactive carbanion for future experiments. Isoprenoid conjugates of nucleoside 5??-diphosphates, which were proposed as either an initiator nucleotide or substrate molecule for in vitro selection of prenyl-transferase ribozyme were prepared by one step nucleophilic displacement reactions. A random DNA pool was constructed for selection of a ketosynthase ribozyme. A substrate bearing a biotin tag was prepared by one-step conjugation. Hig-tagged T7 RNA polymerase was expressed and purified for a large scale transcription reaction. In vitro transcription of the random DNA pool with a 5??-thiol modified GMP analogue as an initiator nucleotide produced a thiol-modified random RNA library.

Many different experimental approaches were attempted to achieve carbon-carbon bond formation by nucleic acid template-directed reactions and ribozyme catalysis as potential lipid synthesizing machineries in the RNA world. A novel biomimetic condition for decarboxylative Claisen condensation in polyketide biosynthesis was discovered. The reaction of a malonic acid half oxyester with a Nhydroxysuccinmidyl ester forming reagent resulted in self-condensation to provide the corresponding 1,3-acetonedicarboxylic acid diester in the absence of a divalent metal chelator or a coordinating solvent. The decarboxylative Claisen condensation of malonyl adenosine using a poly-U template in solution or with immobilized poly-U was attempted. Various analytical methods demonstrated that malonyl adenosine underwent an exclusive hydrolysis reaction instead of condensation in the given conditions. Similar results were observed for the reaction of malonyl-CoA with acetyl-CoA on poly-U templates. No evidence for the decarboxylative Claisen condensation was observed by a DNA-templated system although a double helical structure of DNA duplex was proven to facilitate a bimolecular reaction by offering a favorable proximity effect. Therefore, it seems that the unsuccessful condensation resulted not from the bad template effect but from the intrinsic properties of the decarboxylative Claisen condensation reaction itself. Two tRNA molecules loaded with a malonic acid were prepared by ligation of truncated tRNAs with malonylated dinucletides. Our initial attempts to probe carbon-carbon bond formation by subjecting malonylated tRNAs to the in vitro translational machinery were not successful. Novel carbon isosteres of α-amino acids are suggested as a potential source of a more stable and reactive carbanion for future experiments. Isoprenoid conjugates of nucleoside 5??-diphosphates, which were proposed as either an initiator nucleotide or substrate molecule for in vitro selection of prenyl-transferase ribozyme were prepared by one step nucleophilic displacement reactions. A random DNA pool was constructed for selection of a ketosynthase ribozyme. A substrate bearing a biotin tag was prepared by one-step conjugation. Hig-tagged T7 RNA polymerase was expressed and purified for a large scale transcription reaction. In vitro transcription of the random DNA pool with a 5??-thiol modified GMP analogue as an initiator nucleotide produced a thiol-modified random RNA library.

A subset of mitochondrial group II introns of flowering plants has, over evolutionary time, lost characteristic features and employs unconventional splicing pathways. Given the potential impact of cold treatment on RNA folding, as well as on enzymatic activity and import of nuclear-encoded splicing machinery, I have examined the physical excised forms of aberrant introns from wheat embryos subjected to 4oC. My findings suggest a shift in biochemistry with cold treatment to novel splicing pathways that generate heterogeneous in vivo circularized forms for nad1 intron 2, nad2 intron 1 and the cox2 intron, in contrast to predominantly linear excised intron forms at room temperature. Interestingly, the highly degenerate nad1 intron 1, which due to DNA rearrangement has been broken into two halves that interact for splicing in trans, is excised exclusively by first-step hydrolysis at room temperature and under cold treatment. In this case, splicing culminates in two distinct linear half introns that appears correlated with an unusual 5’ terminal insert. This represents the first in vivo demonstration of hydrolytic trans-splicing. Based on northern analysis, cold treatment was further associated with reduced splicing efficiency for all introns surveyed. Moreover, study of precursor transcripts of the nad1a-intron 1a locus suggests the efficiency of end-maturation, including processing of the cotranscribed tRNA-Pro gene, is also reduced in the cold. My findings demonstrate a temperature-sensitivity of transcript maturation, particularly for RNA splicing, providing new insight into the impact of cold growth conditions on plant mitochondrial gene expression.