A reverse transcription reaction allows the production of complementary DNA (cDNA) using an RNA template and relies on polymerases displaying reverse transcriptase (RT) activity. This process, with major applications in both research and in medical diagnostics, is often limited by the nature of the RTs available. RNA secondary structure can prove problematic where mesophilic retroviral RTs are used while the alternative approach, using thermophilic DNA polymerases with RT activity, often results in error-prone cDNA production. <br /> This project recognised the need to study other possible sources of thermophilic RTs and outlines the study of four previously uncharacterised Group II Intronencoded proteins (IEP), with RT domains, from thermophilic bacteria. While cloning of the IEP genes and their expression on a small scale proved successful, difficulties were encountered when attempting purification. Despite a lack of overall purity, samples containing IEPs from Thermosinus carboxydivorans and Petrotoga mobilis were shown to have RT activity but characterisation of these IEPs was not carried out. However, an IEP from Bacillus caldovelox proved to be an excellent candidate for characterisation as successful purification was achieved. Enzyme engineering was also performed, fusing a Sac7d domain onto the C-terminus of this protein. These enzymes were shown to have optimum RT activity at 54ºC with activity still being displayed at 76ºC. Other studies on these enzymes showed that, unlike the retroviral RTs, the IEPs displayed no DNA-dependent DNA polymerase activity. The Sac7d fusion protein was also studied in terms of possible enhancements to the RT activity of an IEP. However, preliminary studies showed that, although this domain did not prove to be detrimental to the enzyme, it had little effect on improving the processivity of the RTs. <br /> Although this class of RT looks promising in terms of use as an alternative thermophilic RT, the IEPs studied in this report did incur major limitations during cDNA synthesis, which included lower than expected optimum reaction temperatures, very low fidelity and an inability to synthesise cDNA using complex RNA templates.

Mobile group II introns are retroelements that insert site-specifically into double-stranded DNA sites by a process called retrohoming. Retrohoming activity rests in a ribonucleoprotein (RNP) complex that contains an intron-encoded protein (IEP) and the excised intron RNA. The intron RNA uses its ribozyme activity to reverse splice into the top strand of the DNA target site, while the IEP cleaves the bottom DNA strand and reverse transcribes the inserted intron. My dissertation focuses on the Lactococcus lactis Ll.LtrB group II intron and its IEP, denoted LtrA. First, I investigated the ability of microinjected Ll.LtrB RNPs to retrohome into plasmid target sites in Drosophila melanogaster precellular blastoderm stage embryos. I found that injection of extra Mg2+ into the embryo was crucial for efficient retrohoming. Next, I compared retrohoming of linear and lariat forms of the intron RNP. Unlike lariat RNPs, retrohoming products of linear intron RNPs displayed heterogeneity at the 5’-intron insertion junction, including 5’-exon resection, intron truncation, and/or repair at regions of microhomology. To investigate whether these junctions result from cDNA ligation by non-homologous end-joining (NHEJ), I analyzed retrohoming of linear and lariat intron RNPs in D. melanogaster embryos with null mutations in the NHEJ genes lig4 and ku70, as well as the DNA repair polymerase polQ. I found that null mutations in each gene decreased retrohoming of linear compared to lariat intron RNPs. To determine whether novel activities of the LtrA protein contributed to the linear intron retrohoming 5’ junctions, I assayed the polymerase, non-templated nucleotide addition and template-switching activities of LtrA on oligonucleotide substrates mimicking the 5’-intron insertion junction in vitro. Although LtrA efficiently template switched to 5’-exon DNA substrates, the junctions produced differed from those observed in vivo, indicating that template switching is not a significant alternative to NHEJ in vivo. Finally, I designed and constructed retargeted Ll.LtrB RNPs to site-specifically insert into endogenous chromosomal DNA sites in D. melanogaster. I obtained intron integration efficiencies into chromosomal targets up to 0.4% in embryos and 0.021% in adult flies. These studies expand the utility of group II intron RNPs as gene targeting tools in model eukaryotic organisms. / text