ATP dependent chromatin remodelers use the energy from ATP hydrolysis to move, disassemble or alter the composition of nucleosomes. Though all remodelers share a conserved ATP hydrolysis and DNA translocase domain, their biochemical actions and in-vivo characteristics differ because of their subunits and accessory domains in the catalytic subunit that regulate its activity. Understanding how these domains contribute to remodeler regulation in terms of substrate interaction and regulation of the catalytic subunit is therefore important to understanding what causes a remodeler to behave differently, and what are the mechanistic underpinnings of such behavior. In this study we have addressed these questions using the SWI/SNF remodeler from budding yeast (Saccharomyces cerevisiae) to explore how different remodelers compare to SWI/SNF in terms of nucleosome interaction. Using a chemical based histone – remodeler photo-crosslinking and labeling approach, we show that different remodelers contact nucleosomes in patterns unique to their functions, and even remodelers that belong to the same family interact with nucleosomes in a unique manner to accomplish their respective remodeling results. In addition we delineate the functions of the AT hook motifs in the catalytic subunit of SWI/SNF using in-vitro and in-vivo techniques. We demonstrate the necessity of the regulatory action of the motif in the context of SWI/SNF remodeling due to its requirement for efficient ATP hydrolysis by the catalytic domain and therefore efficient remodeling. We also demonstrate for the first time that SWI/SNF in yeast is involved in transcriptional repression with evidence that the AT hook alters SWI/SNF activity at particular genomic regions. Regulation of SWI/SNF activity is an increasingly important topic of study, with mutations that cause SWI/SNF dysfunction being implicated in a large number of cancers and neurological diseases. We attempt to find out the biochemical implications of mutations in the catalytic, SnAC and AT hook motifs with respect to SWI/SNF activity. Taken together, this study provides an insight into some of the different mechanisms in which remodelers are regulated using budding yeast as a model system.

Chromatin remodelling proteins support a variety of cellular functions and utilise the energy from ATP hydrolysis to either reposition or evict nucleosomes. One such protein, Lymphoid specific helicase (LSH), regulates DNA methylation in mammalian cells cooperatively with DNA Methyltransferase 3B (DNMT3B) through binding of the N-terminal domain of LSH. The correct functioning of LSH is essential for heterochromatin formation, with a knockout of LSH causing perinatal lethality or severe developmental abnormalities. There is little biochemical data and no structural data on LSH. Therefore, we aim to determine the structural characteristics and regulatory mechanism of LSH in vitro. LSH was expressed in an optimised insect cell system which increased protein yield 25-fold with greater than 95% purity. LSH is monomeric with increased thermal stability upon ATP or ADP binding. Full length LSH could not be crystallised therefore a core ATPase region of LSH missing the N-terminal domain was identified through limited proteolysis. This also provided evidence the N-terminal domain of LSH is disordered, which was proven through biophysical characterisation of LSH1-176. Expression of the LSH ATPase region was weak and the protein was unstable; suggesting the N-terminal domain of LSH is required for LSH stability. Therefore, complementary structural methods were used to study LSH. Crosslinking mass-spectrometry revealed the N and C termini are in close proximity, suggesting flexible linking regions, which was supported by limited proteolysis experiments. Negative staining Electron Microscopy defined LSH as a tri-lobal and elongated structure which could harbour the ATPase region in the two spherical lobes. 3D modelling of SAXS data obtained of LSH was in agreement with EM data. To understand molecular mechanisms of LSH, functional studies investigating LSH:DNA and LSH:DNMT3B interactions were performed. LSH had a KD for dsDNA of 0.4 μM in solution. LSH does not bind ssDNA nor does it have a greater affinity for methylated dsDNA. LSH was found to bind the dsDNA overhangs of nucleosomes but not to core nucleosomes, suggesting LSH solely interacts with DNA in chromatin and not histones. A stable complex of LSH:DNMT3B could not be achieved in vitro, however, other components for complex formation may have been missing. This study has improved our understanding of LSH structure, biophysical properties and its biochemical interaction with DNA and nucleosomes. This study has laid the foundations for the structural investigations of a LSH:nucleosome and potentially a LSH:DNMT3B complex in vitro to gain a greater understanding of how functional domains of LSH regulates its enzymatic function.

<p dir="ltr">H3K27me3 is a transcriptional repressive epigenetic mark that plays vital roles in many biological processes in <i>Arabidopsis thaliana</i>. A number of biochemical and functional characterizations of PKL, an ATP-dependent CHD chromatin remodeler, suggest that PKL contributes to maintain the homeostasis of H3K27me3. To identify other factors that act with PKL together to contribute to the homeostasis of H3K27me3, we undertook an EMS-mutagenesis screen for <i>pkl</i>-associated phenotypes. This genetic screen suggests that PKL may contribute to maintaining the homeostasis of H3K27me3 in an H2A.Z associated or a Mediator associated pathway.</p><p dir="ltr">Here, we took advantage of a combined genetic and bioinformatic method to characterize the contribution of PKL in these two pathways as described above. Our analysis revealed a robust genetic interaction between <i>HTA9</i>, <i>HTA11</i>, and <i>PKL</i> in maintaining proper H2A.Z distribution and enrichment of H3K27me3. In addition, the characterization also sheds light on unexpected roles of PKL in promoting the homeostasis of H3K4me3 and acting with histone demethylases to promote removal of H3K27me3 in an H2A.Z dependent manner. Furthermore, our result also raised the possibility that the tail module of the Mediator complex also plays a critical role in the homeostasis of H3K27me3. While we were examining <i>PKL</i>-dependent chromatin features, we largely optimized the protocol for preparation ChIP-seq samples and libraries and implemented a gene-centric ChIP-seq bioinformatics pipeline for providing robust analysis.</p><p dir="ltr">Ultimately, the work presented in this thesis highlights several divergent pathways that PKL contributes to maintain chromatin homeostasis. By and large, the combined observation from this thesis advances our knowledge of how PKL interacts with other chromatin-associated machineries together to maintain proper epigenetic states and promote other more emergent DNA-templated processes, including replication and transcription.</p>