|dc.description.abstract||Transfer RNA (tRNA) is an ancient molecule with a myriad of functions in modern biology; however, its pre-eminent role is in the transfer of individual amino acids to the growing polypeptide chain in protein synthesis. The weight of both theoretical and experimental evidence points to a hairpin-duplication origin for tRNA, with Di Giulio arguing that the canonical intron insertion position in the anticodon loop marks the point at which the two hairpins were originally joined. I have discovered a molecular ‘signature’ in the anticodon loop sequences of contemporary tRNAsGly: a highly conserved CCA sequence that suggests the first anticodon (NCC for glycine) was derived from the 3'-terminal CCA of the upstream hairpin, and the first tRNA was tRNAGly. The relative lack of post- transcriptional modification in contemporary tRNAsGly suggests that this tRNA could have formed the necessary interactions that form the basis of the genetic code prior to the advent of protein modification enzymes. Such a scenario would appear to fit Crick’s frozen accident model for the origin of the genetic code, but in fact the presence of this particular sequence in the anticodon loop of the first tRNA may have been instrumental in the subsequent evolution of coded protein synthesis.
I have hypothesized that the advent of the single-stranded anticodon loop allowed tRNAs to interact with single-stranded (regions of) RNAs in the RNA world, and these serendipitous binding partners bound to pairs of anticodon loops held in juxtaposition on an ancestral peptidyl transferase ribozyme catalysing noncoded peptide synthesis, the predecessor to the large ribosomal subunit RNA. This interaction would have tethered the two tRNAs, enhancing their binding to the peptidyl transferase and increasing the rate of peptide synthesis. I believe such a system evolved naturally into a system of coded protein synthesis, driven by selection for enhancement of synthesis of specific peptides. A similar model has been discussed very briefly by Christian de Duve in Blueprint for a Cell, and is supported by isolated experiments that demonstrate enhanced binding of tRNA to the Escherichia coli 50S ribosomal subunit (that contains the peptidyl transfer centre) in the presence of cognate oligonucleotides. According to my model, the small ribosomal subunit RNA evolved later as a decoding hairpin that functioned in trans, exerting control over the interaction between tRNAs and their binding partners (proto-mRNAs).
I have also utilized tRNAGly from E. coli as a starting material for the creation of tRNA variants able to oligomerize or self-assemble, with the dual aims of exploring the possible role of self-assembly in the evolution of tRNA and for producing RNA building blocks that will form higher oligomers, ideally 3D arrays, at neutral pH, with potential nanotechnology applications. Site-directed mutagenesis of tRNAGly was used to create tRNA variants with additional complementarity in the T loop and other single-stranded regions, with the aim of enhancing their ability to oligomerize. A tRNA variant with a self- complementary GGGCCC sequence in the T loop formed dimers rapidly at neutral pH inducible by the addition of Mg2+ and/or spermine. However, the introduction of multiple self-complementary sequences resulted in the conversion of the standard cloverleaf into alternative secondary structures due to intramolecular base-pairing interactions, and no higher oligomers were formed. Using oligonucleotide inhibition and nucleotide substitution, a tRNA variant possessing a non-standard elongated structure is proposed to form dimers through single-stranded sequences that occur in stems in the standard cloverleaf structure.||