The induction and repair of DNA damage has been shown to occur heterogeneously throughout the mammalian genome. As a consequence, analysis of these parameters at a global genome level may not reflect important gene-level events. Few techniques have been established to explore quantitatively gene-specific DNA damage and repair. Most of these are PCR-based assays and are relatively insensitive, relying on decreased PCR amplification arising from damage in template DNA. In this study, a quantitative assay that combines specific immunocapture of damaged DNA by an antiserum specific for thymine dimers (IgG479), with PCR amplification of a 149bp fragment of the human H-ras proto-oncogene was established. Quantification of DNA damage was based upon proportionality between the amount of the PCR product and the initial amount of damage. Detection of thymine dimers was possible with nanogram amounts of genomic DNA and increased in a linear, dose-responsive manner. Using this assay, gene-level induction of thymine dimers was shown to be directly proportional to levels induced in the global genome of UVC-and-UVB-irradiated genomic DNA, as measured by gas chromatography-mass spectrometry and enzyme-linked immunosorbent assay, respectively. This result suggests that global damage assessments do indeed reflect gene-level events. However, preferential repair of UVB-induced T < > T from the human H-ras proto-oncogene than the overall genome of human keratinocyte cells was detected. These findings demonstrate the suitability of this approach to the detection of UVR-induced DNA damage and repair at the level of individual genes.

SLPI and elafin are members of the whey acidic protein (WAP) four disulphide core (WFDC} family of proteins and have multiple contributions to innate host defence including inhibition of neutrophil serine proteases, antimicrobial properties and inhibition of the inflammatory response to LPS. This research aimed to investigate novel functions of WAP2 and eppin, two lesser characterised WFDC proteins that are found expressed in the lung. Recombinant expression and purification of WAP2 were first optimised in E. coli for use in a range of functional assays. Recombinant WAP2 inhibited cathepsin G protease activity. Monocytic cells pre-treated with recombinant WAP2 before LPS stimulation produced significantly lower levels of IL-8 and MCP-l compared to cells stimulated with LPS alone. Recombinant WAP2 became conjugated to fibronectin in a transglutaminase-mediated reaction and retained antiprotease activity. WAP2 was detected at variable levels in bronchoalveolar lavage fluid from cystic fibrosis patients. WAP2 and eppin were both shown by immunohistochemistry to be expressed in epithelial and immune cells of the lung. Eppin exerted potent antibacterial effects against the lung-colonising bacteria P. aeruginosa and S. aureus, in contrast to WAP2 which was ineffective. Together these results suggest a role for both WAP2 and eppin in the innate host response. Further characterisation is required to fully understand the range and implications of their physiological functions.

For more than 20 years, a model for the colonisation ofIsland Southeast Asia (ISEA), involving as two waves of dispersal, has assumed the ~status of near-consensus. The first colonisation of the region by modem humans occurred around 50,000 years ago, but according to the 'Out of Taiwan' model, these populations were assimilated or replaced in the mid-Holocene by rice agriculturists from China, who moved to Taiwan and from there into ISEA and later the Pacific. This theory originated in the analysis of the tree of the Austronesian language family, spoken by most of these populations. In this thesis, mitochondrial DNA variation was analysed using a phylogeographic approach in order to test this hypothesis against other proposed models for the colonization of ISEA. Two improved analytical tools were employed to do this. One was an upgrade of the previously used founder analysis approach, and the second a reassessment of human mitochondrial DNA clock. I developed a time-dependent mutation rate that allows for the effect of purifying selection on the more recent nodes of the human complete mitochondrial tree (control region as well as coding region), increasing both the accuracy and precision of the age estimates. The founder analysis revealed that the Late Pleistocene and Early Holocene were the key periods that shaped the mitochondrial diversity of ISEA, possibly due to events related to climate change and sea-level rise, and that the mid-Holocene Neolithic signal traces mainly to Mainland Southeast Asia, rather than Taiwan. The analysis of complete mitochondrial geno~es suggested a first arrival of modem humans 50,000-60,000 years ago, confmned the importance of sea-level rises in the region, and highlighted the significance of interaction networks in establishing current genetic patterns in ISEA, the Pacific and Taiwan. The analysis offers little support for 'the 'Out of Taiwan' model, pointing to a more complex demographic history for the region in which climatic change appears to have played a crucial role.

Eukaryotic genomes are highly organized and packaged into chromatin, a complex structure formed of proteins and DNA, in which the basic repeating unit is the nucleosome. Chromatin can be arranged in condensed or relaxed structures influencing accessibility of proteins that regulate transcription, replication, recombination and repair. One class of transiently chromatin-associated proteins is the High-Mobility Group (HMG) protein family. HMG proteins are subdivided into three subgroups: HMGA, HMGB and HMGN. HMGN1, the subject of this study, is a prominent member of the HMGN (High-Mobility Group Nucleosome-binding) protein family, the only HMG proteins that specifically binds to the nucleosomes. HMGNs are maintained in dynamic balance between nucleosome-associated and nucleosome-free pools. Regulation of chromatin involves several enzymatic activities that modify specific residues on chromatin proteins, which may influence these interactions. While associated with nucleosomes, HMGNs can interfere with some modifications of histone tails. Modification of HMGN1 on specific residues and post-translational modification (PTM) of histones are concomitantly regulated by the complex signalling networks associated with the induction of immediate early (IE) genes. Induction of IE genes is associated with phosphorylation of HMGN1 which has been suggested to increase the rate of dissociation of HMGN1 from the nucleosome, thus allowing access and modification of histone tails. My research has been focused on characterizing HMGN1 isoforms present in different cellular compartments and at different time-points during IE gene induction with various stimuli, including epidermal growth factor (EGF), anisomycin (An) and 12-O-tetradecanoylphorbol-13-acetate (TPA). Furthermore, I investigated the localization of HMGN1 within the nucleus and at specific IE gene loci, especially at sites where post-translationally modified histones are localised. In my analysis only the phosphorylation at serine 6 of HMGN1 shows a correlation with gene induction. Analysis of DNA sequences from chromatin immunoprecipitation (ChIP) has shown that HMGN1 is present at equal levels in active and inactive genes. It appears that HMGN1 localization on DNA is not dictated by a particular preference for any gene elements such as promoters, exons, introns or gene termination sequences.

Alternative splicing (AS) is a common post-transcriptional process in eukaryotic organisms, by which multiple distinct functional transcripts are produced from a single gene. Because of its potential role in expanding transcript diversity, interest in alternative splicing has been increasing over the last decade, ever since the release of the human genome draft showed it contained little more than the number of genes of a worm. Although recent studies have shown that 94% human multi-exon genes undergo AS while aberrant AS may cause disease or cancer, evolution of AS in eukaryotic genomes remains largely unexplored mainly due to the lack of comparable AS estimates. In this thesis I built a Eukaryote Comprehensive & Comparable Alternative Splicing Events Database (ECCASED) based on the analyses of over 30 million Expressed Sequence Tag (ESTs) for 114 eukaryotic genomes, including protists (22), plants (20), fungi (23), metazoan (non-vertebrates, 29) and vertebrates (20). Using this database, I addressed two main questions: 1) How does alternative splicing relate to gene duplication (GD) as an alternative mechanism to increase transcript diversity? and 2) What is the contribution of alternative splicing to eukaryote transcript diversity? I found that the previous “interchangeable model” of AS and gene duplication is a by-product of an existing relation between gene expression breadth, AS and gene family size. I also show that alternative splicing has played a key role in the expansion of transcript diversity and that this expansion is the best predictor reported to date of organisms complexity assayed as number of cell types. In addition, by comparing alternative splicing patterns in cancer and normal transcript libraries I found that cancer derived transcript libraries have increased levels of “noisy splicing”.