One of the most important outstanding questions in biology involves the precise spatial and temporal regulation of gene activity, which enables different cell types to express the specific set of genes required for their function and is therefore a cornerstone for complex biological life. Cis-regulatory elements, such as gene promoters and enhancers, play a key role in controlling gene activity. These elements physically interact with the genes they regulate within structural chromatin domains. The organisation of chromosomes into these domains is critical for specific regulation of gene expression and disruption of these structures underlies common human disease. However, it is not understood how chromatin domains form, how interactions between the cis-regulatory elements contained within them are established, or how such interactions influence gene expression. The major hurdles in addressing these questions are to determine chromatin structures with high resolution and sensitivity and to examine their dynamic nature within single cells. To overcome these, I have developed Tri-C, a new chromosome conformation capture assay that can analyse concurrent chromatin interactions at single alleles at high resolution. By combining Tri-C with conventional chromosome conformation capture techniques, I have analysed the three-dimensional regulatory landscapes of the well-characterised murine globin loci at unprecedented depth. Additionally, to examine the roles of cis-regulatory elements in establishing chromatin architecture, I have analysed how engineered deletions in enhancers and CTCF-binding elements in the murine alpha-globin locus disrupt its chromatin landscape. These analyses reveal that the chromatin domains containing the globin genes represent compartmentalised structures, which are delimited by CTCF boundaries. The heterogeneity of interactions in these domains between individual cells is indicative for a dynamic process of loop extrusion underlying their formation. Within chromatin domains, preferential structures are formed in which multiple enhancers and promoters interact simultaneously. These complexes provide a structural basis for understanding how multiple cis-regulatory elements cooperate to establish robust regulation of gene expression. Importantly, these complex, tissue-specific structures, cannot be explained by loop extrusion alone and indicate other, independent mechanisms contributing to chromosome architecture, likely involving interactions mediated by multi-protein complexes. Together, these analyses show that the current view of genome organisation, in which chromosomes are organised by stable loops and domains that self-assemble into hierarchical structures, is not correct. Rather, chromatin architecture reflects a complex interplay between distinct molecular mechanisms contributing to the formation of regulatory landscapes that facilitate precise, robust control of gene expression.

Meiosis is a specialized form of cell division where homologous chromosomes are segregated in meiosis I before sister chromatids are segregated in meiosis II. To establish this pattern, a number of changes to the mitotic chromosome segregation machinery are put in place. Firstly, sister kinetochores orient towards the same pole in meiosis I (mono-orientation). Secondly, homologue recombination creates chiasmata, which link homologues together. And thirdly, cohesin, the molecule that holds sister chromatids together, is cleaved in a step-wise manner. This is achieved because the Shugoshin (Sgo1) protein recruits protein phosphatase 2A (PP2A) to centromeres to counteract cohesin phosphorylation, which is required for its cleavage. The work presented here has investigated two critical aspects of cohesin protection: firstly, how cohesin protection is deactivated in meiosis II and, secondly, how a meiosis-specific protein called Spo13 helps to set up cohesin protection in meiosis I. Previously, our lab had shown that Sgo1 is removed from chromosomes when sister chromatids come under tension during mitosis. I therefore sought to investigate whether sister kinetochore mono-orientation allows Sgo1 to stay on centromeres during meiosis I and carry out its protective function. To this end, I modified meiosis I chromosomes to lack both chiasmata and mono-oriented kinetochores. Under these conditions, where sister chromatids are forced to be under tension in metaphase I, Sgo1 is undetectable on chromosomes. As a consequence, centromeric cohesin is largely lost in anaphase I leading to the premature separation of sister chromatids in a fraction of cells. Since mono-orientation of sister kinetochores is exclusive to meiosis I, these findings suggest that Sgo1 localisation is influenced by sister kinetochore tension in both mitosis and meiosis. Therefore, our findings suggest a mechanism that could contribute to the deprotection of cohesin in meiosis II. However, loss of cohesin protection upon bi-orientation is not complete, suggesting that other factors are involved in the efficient protection and deprotection of cohesin. One such factor is the meiosis-specific protein Spo13, which had previously been shown to be required for cohesin protection as well as kinetochore monoorientation. Although it had been suggested that Spo13 regulates Sgo1 recruitment to centromeres, I could not find any evidence to support a loss of Sgo1, or PP2A, in spo13Δ cells. Additionally, even when Sgo1 is stabilised and clearly visible in anaphase I of spo13Δ mutants, pericentromeric cohesion is still defective. Therefore, I investigated the effect that polo kinase Cdc5, an interactor of Spo13, has on Sgo1. While cellular Sgo1 levels are increased in response to Cdc5 loss, this effect seems to be independent of Spo13. However, Spo13 is required for proper levels of Cdc5 at centromeres and the centromeric recruitment of Cdc5 by Spo13 is likely to be functionally important because tethering of Cdc5 to kinetochores rescued the mono-orientation phenotype of spo13Δ cells. In contrast, I found no evidence that the Spo13-Cdc5 interaction is required for cohesin protection. Meiotic overexpression of SPO13 enhances cohesin protection in meiosis I, apparently independent of its robust interaction with Cdc5, and causes increased Sgo1 enrichment at centromeres. This suggested that Spo13 might recruit Sgo1 to cohesin itself to facilitate its protection. Although I could not detect a loss of Sgo1-cohesin interaction in spo13Δ cells, tethering of Sgo1 to cohesin restores pericentromeric Rec8 to spo13Δ mutants in anaphase I. Surprisingly, sister chromatids still segregate in this case, suggesting that pericentromeric cohesion is defective, despite maintenance of Rec8. Furthermore, inhibition of either one of the cohesin kinases, DDK and Hrr25, restores sister chromatid cohesion to spo13Δ cells. Therefore, the findings in this study suggest that Spo13 is at the centre of a complex regulatory network that coordinates cohesin protection and sister chromatid cohesion in meiosis I.

Genomic instability and aneuploidy, which are ubiquitous hallmarks of cancer cells, encompass both structural and numerical chromosome aberrations. Strikingly, cancer cells often display recurrent patterns of aneuploidy which are thought to be contingent on selection pressures within the tumour microenvironment maintaining advantageous karyotypes. However, it is currently unknown if individual chromosomes are intrinsically vulnerable to missegregation, and therefore whether chromosome bias may also contribute to pathological aneuploidy patterns. Moreover, the earliest responses to chromosome missegregation in non-transformed cells, and how these are overcome in cancer, has remained elusive due to the difficult nature of isolating nascent aneuploid cells. Results. Individual chromosomes displayed recurrent patterns of biased missegregation in response to a variety of cellular stresses across cell lines. Likewise, a small subset of chromosomes accounted for a large fraction of segregation errors following one specific mechanism driving aneuploidy. This was supported by the discovery that chromosomes 1 and 2 are strikingly susceptible to the premature loss of sister chromatid cohesion during prolonged prometaphase arrest. Additionally, I have elucidated the arrangement of individual metaphase human chromosomes, highlighting missegregation vulnerabilities occurring at the metaphase plate periphery following nocodazole wash-out. Finally, I have developed a novel system for isolating nascent aneuploid cells, suggesting the earliest transcriptome responses to chromosome missegregation in non-transformed human cells involve ATM and BCL2-mediated apoptosis.

No description available.

The centromere is a chromosomal locus required for accurate segregation of sister chromatids during cell division. They are maintained epigenetically in most eukaryotes, by incorporating the H3 variant CENP-A, and can, in rare instances, change location on the chromosome throughout generations. Centromeres are transcribed, and an active transcription chromatin signature is required for centromere maintenance. For this reason, insight into the nature of this so-called “centrochromatin” is essential for understanding a centromere’s place in the chromosome. The body of work contained in this thesis shows my efforts to understand the centromere in the context of chromatin, revealing interactions and new evidence for repressive chromatin domains with centromere activity, in two different vertebrate models: chicken DT40 cells and human HeLa cells. Centromeres are generally embedded within large domains of heterochromatic repetitive sequences in most eukaryotes, and mapping “centrochromatin” to high-resolution has proven difficult. However, chromosomes 5, 27 and Z of Gallus gallus are not located within repeat arrays, and are fully sequenced. CENP-A distribution on these centromeres has been mapped by ChIP-seq, and I have performed ChIP against selected histone modifications as part of a collaboration. While levels of heterochromatin are naturally quite low in these centromeres, I have shown that repressive polycomb chromatin instead is enriched in these non-repetitive centromeres, suggesting a replacement of one silenced chromatin state with another. Additional mapping of these centromeres showed a pattern of active chromatin marks distinct from that reported for human cells, which exhibited dynamic distribution throughout the cell cycle. Furthermore, conditionally generated neocentromeres in DT40 cells revealed that centrochromatin formation lowers, but does not eliminate, active transcription. To directly study the interaction of polycomb and heterochromatin with centrochromatin, I used a synthetic Human Artificial Chromosome (HAC), which allows for specific conditional targeting of chromatin modification enzymes, allowing manipulation of the underlying chromatin. Enrichment of the polycomb chromatin state on the HAC centromere, by EZH2 tethering, reduced its active transcriptional chromatin signature, but did not impair its actual transcription or mitotic activity. However, direct tethering of polycomb secondary silencing effector PRC1 caused centromere loss, and this effect was mimicked with homologous heterochromatin factors, indicating that centromeres can subsist within repressive chromatin domains, but are lost when direct repression is applied. To understand the contribution of the local repressive heterochromatin to centromere stability, I erased heterochromatin marks from the HAC centromere by tethering JMJD2D (an H3K9me3 demethylase): long-term (but not short-term) heterochromatin loss impaired CENP-A assembly, perturbed mitotic behaviour, and resulted in significant HAC mis-segregation. These results strongly suggest that local heterochromatin is essential to maintain normal CENP-A dynamics and centromere function. Together with previous observations, these data suggest that a repressive chromatin environment contributes to centromere stability, and that centromeres likely have natural mechanisms to maintain their transcriptional activity within such domains.

In order to contain many millions, or even billions of base pairs within every nucleus of a eukaryotic cell, DNA must be extensively packaged. This is achieved by association of DNA with packaging proteins, resulting in the formation of chromatin, which can lead to various degrees of compaction. The most extreme form of compaction is the highly condensed mitotic chromosome, formation of which is necessary for proper resolution and segregation of the genetic material during cell division. However, the exact nature of the structure of chromatin within the mitotic chromosome and the factors which regulate it remain subjects of debate and continued investigation. The hybrid cell line F1.1 presents a unique tool for the study of mitotic chromosome structure. This mouse cell line has been observed to present a distinct chromatin structure in mitosis assembled over a large region of DNA inserted into one of its chromosomes and originating from the fission yeast Schizosaccharomyces pombe. Direct comparison of the structure of this distinct region of chromatin with that of the adjacent endogenous chromatin could provide insight into the nature of mitotic chromosome structure as well as the properties of the chromatin which are influencing this structure. Microscopy and Hi-C analyses showed that the mitotic chromatin organising or "scaffold" proteins are not altered over the region of S. pombe chromatin, but that the amount of chromatin organised around these proteins is diminished. In accordance with the "radial-loop" model of mitotic chromosome structure, we put forward a model whereby the S. pombe chromatin is organised into smaller chromatin loops around a constant organising scaffold. Examination of the histone post-translational modifications over the region of S. pombe chromatin revealed it to be highly heterochromatic, with high levels of H3K9me3 and associated factors such as HP1α and 5meC, and low levels of activating marks. Generation of further mammalian - S. pombe fusion cell lines recapitulated both the distinct mitotic structure and the heterochromatic profile of the inserted S. pombe chromatin. However, insertion of S. pombe DNA into a mouse cell by transfection rather than fusion resulted in a large region of S. pombe DNA that lacked both a distinct structure and heterochromatin. These results suggest that H3K9me3- mediated heterochromatin may influence the structure of chromatin in mitosis, leading to an organisation into smaller chromatin loops than non-heterochromatic regions.

All living organisms need to accurately replicate their genome to survive. Genomic replication occurs in three phases; initiation, elongation, and completion. While initiation and elongation have been extensively characterized, less is known about how replication completes. In Escherichia coli completion occurs at sites where two replication forks converge and is proposed to involve the transiently bypass of the forks, before the overlapping sequences are resected and joined. The reaction requires RecBCD, and involves several other gene products including RecG, ExoI, and SbcDC but can occur independent of recombination or RecA. While several proteins are known to be involved, how they promote this reaction and the intermediates that arise remain uncharacterized. In the first part of this work, I describe the construction of plasmid "mini-chromosomes" containing a bidirectional origin of replication that can be used to examine the intermediates and factors required for the completion reaction. I verify that these substrates can be used to study the completion reaction by demonstrating that these plasmids require completion enzymes to propagate in cells. The completion enzymes are required for plasmids containing two-replisomes, but not one replisome, indicating that the substrate these enzymes act upon in vivo is specifically created when two replication forks converge. Completion events in E. coli are localized to one of the six termination (ter) sequences within the 400-kb terminus region due to the autoregulated action of Tus, which binds to ter and inhibits replication fork progression in an orientation-dependent manner. In the second part of this work, I examine how the presence of ter sequences affect completion on the 2-replisome plasmid. I show that addition of ter sequences modestly decreases the stability of the two-replisome plasmid and that this correlates with higher levels of abnormal, amplified molecules. The results support the idea that ter sites are not essential to completion of DNA replication; similar to what is seen on the chromosome. Rec-B-C-D forms a helicase-nuclease complex that, in addition to completion, is also required for double-strand break repair in E. coli. RecBCD activity is altered upon encountering specific DNA sequences, termed chi, in a manner that promotes crossovers during recombinational processes. In the third part of this work, I demonstrate that the presence of chi in a bidirectional plasmid model promotes the appearance of over-replicated linear molecules and that these products correlate with a reduced stability of the plasmid. The effect appears specific to plasmids containing two replisomes, as chi on the leading or lagging strand of plasmids containing one replisome had no-effect. The observation implies chi promotes a reaction that may encourage further synthesis during the completion reaction, and that at least on the mini-chromosomes substrates, this appears to be a destabilizing force.

This thesis explores the variety of ways in which scientific views of genetic testing are portrayed in the realm of popular culture. As a case study, I have used the identification of the gene for hereditary stomach cancer which occurred in New Zealand in 1998, and was the result of a partnership between the affected whanau and scientists from the University of Otago. Both the empirical and theoretical findings of this project have shown how such accounts are not neutral or transparent. Rather, they are positioned to represent certain values and ideas, and this is even more evident when those affected are Maori. However, considering textual representations of the gene and cancer has revealed the importance of taking into account the fact that these 'things' are also physical and material. I consider the implications of this and consider the ways in which the whanau health workers negotiate the fetishism apparent in biomedicine. Despite its misgivings, biomedicine has immense benefits, some of which the whanau have manipulated and appropriated for their own good, although they do so on their own terms. Despite the many complexities involved in this case study, this is a positive and hopeful story where those involved in the stomach cancer gene project have emerged with improved solutions.

Restriction enzyme, molecular cloning and DNA annealing techniques have been used to study mRNA and DNA coding for the embryonic feather keratins of the chicken and the DNA genomes of coliphages 186 and P2. The coliphage DNAs were used to develop the techniques for application to the keratin system which awaited the availability of appropriate bio - hazard containment facilities before being undertaken. The following results were obtained. 1. Restriction endonuclease cleavage of chick DNA with BamHI, BgïII, EcoRI, or HindIII, fractionation on agarose gels, immobilization on nitrocellulose filters and annealing to DNA complementary to purified 12S mRNA isolated from the developing embryonic feather and coding for embryonic feather keratins, yielded a complex pattern of major and minor bands. These patterns consisted of 4 - 6 major bands and many minor bands. No simple repeat length could be deduced from these patterns, suggesting that keratin - coding DNA is heterogeneous in coding sequences, non - coding sequences or both. 2. Keratin gene expression was shown to be independent of DNA rearrangement, as the complex pattern of restriction fragments was identical in DNA isolated from germ - line tissue ( sperm ) the differentiated feather tissue and somatic tissue not synthesizing keratins ( erythrocytes ). Keratin gene expression must therefore involve the activation of pre - existing control regions in the DNA. 3. The purified 12S mRNA coding for feather keratin was transcribed into double - stranded DNA and individual species isolated by molecular cloning in E. coli. Sequence variation between species was confirmed by restriction enzyme analysis. 4. Preliminary analysis of the cloned species revealed the existence of two distinct groups of species comprising 12S mRNA : Group I ( the more abundant group ) and Group II ( the less abundant ). The fact that filter - bound DNA of individual Group I species bound more 12s cDNA than equal amounts of Group II species DNA and that pure Group I species and total 12S mRNA sequences ( coding for keratins in cell - free translation systems ) annealed to exactly the same complex set of EcoRI, HindIII, or BgïII restricted chick DNA fragments, compels the conclusion that Group I species represent true keratin coding sequences. Group II species annealed to restricted chick DNA fragments which were totally different to those annealing, to either Group I species or total 12S mRNA sequences. Different Group II species appeared to anneal to certain common fragments, suggesting that this less abundant group was comprised of a family of sequence related species and were not simply contaminating mRNA species coding for ' housekeeping ' functions. Their exact nature is at present, however, uncertain. 5. Group I species, the presumptive keratin - coding species, are members of a family of homologous species present in the chick genome. This is demonstrated by the fact that the two Group I species which have been examined so far, shown to be non - identical by restriction analysis, and total 12S mRNA sequences from which they were derived, annealed to the same set of between 20 and 30 BglII, HindIII or EcoRI restricted chick DNA fragments under annealing and washing conditions of low stringency, ( high salt ). Under stringent ( low salt ) washing conditions, however, all except between 1 and 3 of the duplexes formed by these fragments and the Group I species were differentially lost from the filter, indicating that the majority of duplexes were mis - matched and therefore that these multiple copies were homologous and not identical. In addition the two non - identical Group I species annealed to EcoRI generated chick DNA fragments of different sizes under the stringent ( low salt ) washing conditions, demonstrating that differences must exist in the sequence of adjacent non - coding and / or intervening sequences ( should they exist ) for these two species. 6. Although the two Group I species discussed above annealed to different EcoRI generated chick DNA fragments under the stringent ( low salt ) washing conditions, they both annealed under these conditions to a HindIII generated chick DNA fragment of size 3.0 kb. Assuming that this is a single fragment and not two fragments co - electrophoresing by chance, sequences identical to or with very close homology to both of these species lie on the same fragment and are therefore linked in the genome. The exact nature of this linkage and of the extent of gene clustering, should it exist, was not determined. 7. Restriction cleavage maps of coliphages 186 and P2 were determined for the enzymes BamHI, BglII, EcoRI, HindIII, PstI, SaïI, XbaI, and XhoI. These maps were used to analyse four insertion or deletion mutants affecting the major control region of 186. 186ins2 and 186ins3 were shown to be insertions of an IS3 element in the cI. gene and int gene respectively. 186dell and 186del2 were shown to carry the same deletion affecting the cI gene, but 186del2 carried a cryptic insert in the repressor binding site ( operator ). / Thesis (Ph.D.)--Department of Biochemistry, 1979.

Thesis (Ph. D.)--University of Hong Kong, 2007. / Also available online.