Bloom's syndrome is a rare autosomal recessive disorder characterised by an early onset of cancer of many types, erythematous lesions on sun-exposed skin, retarded growth, immunodeficiency and sub- or infertility. Cells from Bloom's syndrome patients have replication defects and an abnormally unstable genome manifested in chromosomal breaks and deletions and in an increased mutation rate. Most characteristically, these cells show elevated levels of sister-chromatid exchanges which probably result from homologous recombination events. Since the cells are not hypersensitive to DNA damaging agents, the defect is unlikely to be in one of the common DNA repair pathways. The gene mutated in Bloom's syndrome, BLM, was cloned in 1995 and found to encode a helicase from the RecQ family. This family is named after its E. coli member, RecQ, and includes at least five human genes. Three of these are mutated in inherited disorders; Bloom's syndrome, Werner's syndrome and Rothmund-Thomson syndrome. In my DPhil project, I have investigated the enzymatic properties of the BLM protein. I have purified the protein in recombinant form and shown that it is a DNA-dependent ATPase and an ATP-dependent helicase with 3'-5' polarity. It binds and unwinds a variety of DNA structures, with a preference for tetraplex (G4)-DNA, Holliday junctions (recombination intermediates) and internal DNA bubbles. Furthermore, it is capable of branch migration, an activity distinct from its helicase activity. BLM forms oligomeric rings with fourfold and sixfold symmetry, both in a cell extract and as purified protein. These results, in combination with the cellular phenotype of Bloom's syndrome and with evidence from the analysis of other RecQ homologues in model organisms such as yeast and E. coli, point to a role for BLM in somatic recombination (recombinational repair). Models for this function are discussed in this thesis.

published_or_final_version / Pathology / Doctoral / Doctor of Philosophy