RNA m5C methyltransferases are a group of enzymes that catalyze the transfer of a methyl group to a cytosine nucleotide of RNA. Only two of these enzymes have been well characterized: Fmu from E. coli and Trm4p from S. cerevisiae. YNLO22c is one of three ORFs identified in S. cerevisiae that have homology with both known and putative RNA m5C methyltransferases, but its encoded protein, YNLO22p, has not been confirmed to have enzyme activity. Verifying that YNLO22c encodes an RNA m5C methyltransferase will require adequate amounts of soluble YNLO22p for enzyme assays. A bacterial expression plasmid for YNLO22c was developed, but the result was insoluble protein. Therefore, several methods known to improve protein solubility were tested to develop a system in which a sufficient amount of soluble YNLO22p could be produced. Results of this study found that coexpression of YNLO22c with chaperone proteins can provide sufficient quantities of soluble YNLO22p. / Department of Biology

Thesis (Ph.D.)--Ohio University, March, 2007. / Includes bibliographical references (leaves 165-175)

Thesis (Ph. D.)--University of Texas at Austin, 2008. / Vita. Includes bibliographical references.

Thesis (Ph.D.)--Ohio University, November, 2007. / Abstract only has been uploaded to OhioLINK. Includes bibliographical references (leaves 174-186)

Thesis (Ph. D.)--University of Texas at Austin, 2004. / Supervisors: Benjamin Kuipers and Robert K. Jansen. Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Texas at Austin, 2008. / Vita. Includes bibliographical references.

Translational efficiency is dictated in part by the availability of charged transfer RNA. Depletion of aminoacylated tRNAs (e.g. during recombinant protein expression) can increase translational errors and associated stress responses. Here, the role of tRNAs as regulators of gene expression was explored through development of synthetic, tRNA-regulated gene circuits, and through an investigation of the impact of tRNA aminoacylation on endogenous gene expression. Synthetic gene circuits initially explored the use of dominant negative alleles of the release factor eRF1 to modulate stop codon readthrough and translationally regulate gene expression. Mutant eRF1 proteins exhibited only a six-fold stimulatory effect on stop codon readthrough. The dominant negative phenotype was rescued partially by overexpression of eRF1, but not eRF3. Ultimately the severity of growth inhibition by these eRF1 alleles limited their utility in synthetic gene circuit design. A novel synthetic circuit was then implemented that utilised TetR interaction with a TetR-inducing peptide in order to control the expression of a suppressor tRNA, and thus a luciferase reporter gene. Using a parameterised mathematical model, the promoter configuration of the circuit was successfully optimised, allowing suppressor tRNAs to regulate the production of luciferase in both feedforward and positive feedback modes of operation. The effects of charged tRNA levels on the global translation network were dissected by regulating the S.cerevisiae glutamine tRNA synthetase gene GLN4 using a tet-off doxycyclineregulated promoter. tRNA synthetase depletion caused the activation of the Gcn4 amino acid starvation response due to accumulation of uncharged glutamine tRNAs. Doxycycline GLN4 shut-off caused increased amino acid production, and decreased ribosome biosynthesis at the transcriptomic and proteomic level, and further physiological changes proposed to result from compromised translation of glutamine-rich regulatory proteins. tRNA overexpression in the GLN4 depletion strain successfully caused altered competition between different isoacceptor tRNA types for their cognate synthetase resource. Together, these results support a growing understanding of tRNA as a key modulator of translation and gene expression in synthetic and natural systems.

The tunneling fold (T-Fold) superfamily is a small superfamily of enzymes found in organisms encompassing all kingdoms of life. Seven members have been identified thus far. Despite sharing a common three-dimensional structure these enzymes perform very diverse chemistries. QueF is a bacterial NADPH-dependent oxidoreductase that catalyzes the reduction of the nitrile group of 7-cyano-7-deazaguanine (preQ0) to a primary amine (preQ1) in the queuosine biosynthetic pathway. Previous work on this enzyme has revealed the mechanism of reaction but the cofactor binding residues remain unknown. The experiments discussed herein aim to elucidate the role of residues lysine 80, lysine 83, and arginine 125 (B. subtilis numbering) in NADPH binding. The biological role of the disulfide bond between the conserved residues cysteine 55 and cysteine 99 observed in several crystal structures is also examined. Characterization of QueF mutants K80A, K83, R125A and R125K revealed lysine 80, lysine 83 and arginine 125 are required for turnover. Further analysis of turnover rates for R125K are consistent with this residue and both lysines being involved in cofactor binding presumably by interacting with the negatively charged phosphate tail of NADPH and are therefore involved in cofactor binding. Based on bond angles and energies, the disulfide bond between Cys55 and Cys99 was characterized as non-structural. Enzyme oxidation assays were consistent with the bond serving to protect QueF against irreversible oxidation of Cys55, which would render the enzyme inactive. This is the only known example of a stress protective mechanism in the Tunneling-fold superfamily. QueF-like is an amidinotransferase found in some species of Crenarchaeota and involved in the biosynthesis of archaeosine-tRNA. The work presented here is focused on the preliminary characterization of this enzyme, including the elucidation of the natural substrate as well as the source of ammonia. The structure of the enzyme was solved and is also discussed. Substrate analysis for QueF-like indicated this enzyme is capable of binding both preQ0 and preQ0-tRNA and reacting to form a thioimide intermediate analogous to QueF but only the latter serves as a substrate for the reaction. This makes QueF-like the first example of a nucleic acid binding enzyme in the Tunneling-fold superfamily. Ammonia, glutamine and asparagine were tested as nitrogen sources and unlike most known amidotransferases, QueF-like can only use free ammonia to produce the archaeosine-tRNA product. The crystal structure of P. calidifontis QueF-like indicates the functional enzyme is a dimer of pentamers pinned together by a large number of salt bridges. The structure presents a high degree of similarity to that of QueF albeit the higher twist of the QueF-like pentamers with respect to QueF results in a more compact structure.

N6-threonylcarbamoyl adenosine (t6A) is a universally conserved tRNA modification found at position 37 of tRNAs which decode ANN codons. Structural studies have implicated its presence as a requirement for the disruption of a U-turn motif in certain tRNAs, leading to the formation of properly structured anticodon stem loop. This structure is proposed to enhance the base pairing between U36 of tRNA and A1 of the codon which aids in translational frame maintenance. Despite significant effort since its discovery in the 1970s the enzymes involved in its biosynthesis remained undiscovered. Bioinformatic analysis identified two proteins as likely candidates for t6A synthesis, YrdC and YgjD. Subsequent gene knockout experiments in yeast were consistent with their involvement in t6A biosynthesis in vivo. Furthermore, clustering between the bacterial genes ygjD, yeaZ and yjeE as well as the identification of a protein interaction network between YgjD, YeaZ, and YjeE suggested that YeaZ and YjeE might be involved in t6A biosynthesis. The genes encoding ygjD, yeaZ, yrdC and yjeE were cloned from E. coli and the recombinant protein was purified. Experiments analyzing the incorporation of [U-14C]-L-threonine and [14C]-bicarbonate (substrates previously indicated in its biosynthesis) into tRNA in the presence of these four proteins demonstrated the first reconstitution of the t6A pathway in vitro. LC-MS analysis verified the formation of t6A, and these proteins were renamed TsaD (YgjD), TsaB (YeaZ), TsaC (YrdC), and TsaE (YjeE). Biochemical characterization of this pathway suggested that the formation of t6A proceeds through an unstable threonylcarbamoyl adenosine monophosphate (TC-AMP) intermediate, which is produced by TsaC from its substrates CO2, L-threonine and ATP. To investigate this reaction in more detail a coupled assay was developed, enabling sensitive detection of turn over. TsaC is a promiscuous enzyme which readily accepts several amino acids as substrates. The formation of t6A from TC-AMP is catalyzed by TsaD, TsaB, and TsaE. Of these three proteins only TsaD is universally conserved suggesting it is the protein catalyzing the transfer of the threonylcarbamoyl moiety to A37 of tRNA. This transfer is not promiscuous as only TC-AMP serves as an efficient substrate for t6A formation. Structural investigation of these proteins are consistent with the formation of a single protein complex potentially alleviating issues with the reactivity and instability of TC-AMP.

The purpose of this study was to investigate human tRNA gene structure, organization, and expression by isolating and analyzing several human transfer RNA genes.