Cytoplasmic dynein 1 is crucial for many cellular processes including endocytosis and cell division. Dynein malfunction can lead to neurodevelopmental and neurodegenerative disease, such as intellectual disability, Charcot-Marie-Tooth disease and spinal muscular atrophy with lower extremity predominance. We formulate, based on physical principles, a mechanical model to describe the stepping behaviour of cytoplasmic dynein walking on microtubules. Unlike previous studies on physical models of this nature, we base our formulation on the whole structure of dynein to include the temporal dynamics of the individual components such as the cargo (for example an endosome or bead), two rings of six ATPase domains associated with diverse cellular activities and the microtubule binding domains. This mathematical framework allows us to examine experimental observations across different species of dynein as well as being able to make predictions (not currently experimentally measured) on the temporal behaviour of the individual components of dynein. Initially, we examine a continuous model using plausible force functions to model the ATP force and binding affinity to the microtubule. Our results show hand-over-hand and shuffling stepping patterns in agreement with experimental observations. We are able to move from a hand-overhand to a shuffling stepping pattern by changing a single parameter. We also explore the effects of multiple motors. Next, we explore stochasticity within the model, modelling the binding of ATP as a random event. Our results reflect experimental observations that dynein walks using a predominantly shuffling stepping pattern. Furthermore, we study the effects of mutated dynein and extend the model to include variable step sizes, backward stepping and dwelling. Independent stepping is studied and the results show that coordinated stepping is needed in order to obtain experimental run lengths.

Aminoglycoside antibiotics are a class of drug used to treat bacterial infections but have the unfortunate side effect of being both oto- and nephro-toxic. Deafness caused by aminoglycoside ototoxicity results from a loss of sensory hair cells from the inner ear. In vitro, two early effects of aminoglycoside exposure can be observed. First, membrane blebs are formed around the perimeter of the hair-cell apical surface. Secondly phosphatidylserine (PS), an aminophospholipid that is normally restricted to the inner leaflet of the plasma membrane, flops to the outer leaflet. This membrane damage occurs rapidly, within 90-120 seconds of drug exposure and can be completely reversed. The aim of this thesis was to determine the molecular mechanisms underlying damage repair in sensory hair cells recovering from aminoglycoside damage. TEM studies using cationic ferritin as a tracer indicates the repair process involves membrane internalisation, but recovery cannot be blocked by inhibitors of macropinocytosis, the clathrin-independent carrier (CLIC) pathway, PI3 kinase, PKC, Pak1 or of the clathrin-coated pit pathway. Damage repair is, however, prevented by the actin stabiliser jasplakinolide and the inhibitor of Protein kinase A, H-89. In addition, the CLIC pathway inhibitor EIPA has been uncovered as a reversible blocker of aminoglycoside entry into hair cells.

No description available.

The aim of this project was to investigate the effects of a novel mouse cytoplasmic dynein mutation; Abnormal rear leg (Arl). Cytoplasmic dynein is a microtubule (MT) based motor protein important for diverse cellular processes including Golgi maintenance and retrograde transport of organelles. Arl is a mouse point mutation in the heavy chain subunit of dynein (Dync1h1). Homozygous Dync1h1Arl/Arl die at embryonic day 10. Dync1h1Arl/+ heterozygotes have a normal life span, but exhibit abnormal gait and hindlimb clasping during tail suspension, typical of neuronal dysfunction. Protein purification from wildtype and heterozygous brain tissue showed increased MT binding in Dync1h1Arl/+ compared to wildtype. Delayed endosomal trafficking was observed in EGF stimulated Dync1h1Arl/+ mouse embryonic fibroblasts (MEFs) compared to wildtype, in both fixed cells and using live cell imaging. Similarly, a delay in the reassembly of the Golgi complex after disruption with a MT depolymerisation agent, nocodazole, was observed in Dync1h1Arl/+ MEFs compared to wildtype. In addition, the Golgi complex was observed as being structurally perturbed in Dync1h1Arl/+ lumbar spinal cord neurons using transmission electron microscopy (TEM) compared to the wildtype. TEM also revealed that the mitochondria were structurally perturbed in Dync1h1Arl/+ lumbar spinal cord neurons compared to wildtype, and O2 consumption assays investigating their function showed the Dync1h1Arl/+ mitochondria to have increased respiration rates compared to wildtype. Thus, these data highlight the Arl mouse as an invaluable model for studying the mechanism of dynein function and the subsequent outcomes when they are compromised.

Maintenance of the intracellular NADH/NAD+ redox poise is vital for energy generation in cells. Gram-positive bacteria, including the antibiotic-producing organism, Streptomyces coelicolor, have evolved a regulatory protein Rex that both senses this ratio and mediates an adaptive response to changes in it. Rex is a dimeric redox-sensitive transcriptional repressor. It is capable of binding to both NAD+ and NADH, although only NADH is an effector, causing dissociation of the protein from operator (ROP) sites. As NADH levels rise during oxygen limitation Rex dissociates from its target genes allowing expression, which helps to restore the NADH/NAD+ ratio. Microarray-based expression studies had suggested that Rex regulated only a small number of genes. In this work, however, ChIP-on-chip analyses revealed 38 genes that are potential regulon members. Analysis of the Rex binding sites in S. coelicolor revealed new insights into the mode of binding and show that Rex can bind with low affinity to incomplete half sites. This work also focused on characterising two key Rex targets, ndh and nuoA-N, that encode non-proton-translocating and proton translocating NADH dehydrogenases, respectively. Whereas nuoAN is not essential and was not expressed in liquid media, ndh was essential for growth. Depletion of NDH from growing cells led to the induction of Rex target genes confirming that ndh and Rex play key roles in maintaining redox homeostasis. Structure-based dissection of Rex, via a close homologue in Thermus aquaticus, identified a key interaction between the NADH- and DNAbinding domains of Rex. An R29-D203' salt-bridge, that traverses the NADH binding and DNA binding domains of Rex, appeared to stabilise the DNA-bound form of Rex, but is ‘broken' in the presence of NADH. In the NADH-bound form of Rex, D203 alternatively interacts with Y111, which in turn interacts with the nicotinamide ring of NADH. In order to assess the importance of individual subunits in the dimeric Rex, a single-chain derivative was constructed and the NADH binding and DNA binding domains individually disrupted.

Culturing C2C12 myoblasts and myotubes with a combination of LPS, TNF-α, IFN-γ and IL1β for 18 hours was used to determine the effects of endotoxic shock on possible causes of the dysregulation of glucose homeostasis associated with the syndrome. The in vitro model was confirmed by the significant production of NO in both myoblasts and myotubes following treatment. The treatment resulted in significantly different results between both myocyte preparations with regards to the regulation of glycogen synthesis. In the myoblasts, the treatment significantly increased myoblast glycogen synthesis, in a NO-independent manner, as seen by the inclusion of the NO synthase inhibitor L-NAME. This stimulation was unlikely to be due to a change in either GS or Phosphorylase activity. However it may have been caused by a significant increase in glucose transport induced by the treatment. This latter increase was also NO-independent, as well as not requiring reactive oxygen species. Insulin-induced myoblast protein synthesis was impaired by the treatment, which is likely due to an impairment of insulin-stimulated ERK1/2 phosphorylation. In the myotubes the case was different, as the treatment significantly reduced glycogen synthesis in a NO-dependent manner. This correlated with a NO-dependent increase in GS phosphorylation, indicating it was less active, however measurements of GS fractional activity failed to confirm this. Insulin stimulation of myotube glycogen synthesis was impaired by the treatment in a NO-independent manner, which may have involved an impairment of the insulin signal to ERK1/2. However the latter impairment was NO-dependent, suggesting other contributory mechanisms. Endotoxic treatment significantly increased myoblast protein content, but failed to do so in myotubes. This effect in the myoblasts may be explained by a significant increase in protein synthesis between 6 and 12 hours of treatment. None of the effects observed in the study were due to the treatment compromising cell viability.

Meiotic recombination is initiated by self-inflicted DNA breaks and primarily involves homologous chromosomes, whereas mitotic recombination involves sister chromatids. Whilst the mitotic recombinase Rad51 exists during meiosis, its activity is suppressed in favour of the meiosis-specific recombinase, Dmc1, thus establishing a meiosis-specific mode of homologous recombination (HR). A key contributor to the suppression of Rad51 activity is the synaptonemal complex (SC), a meiosis-specific chromosomal structure that adheres homologous chromosomes along their entire lengths. Here, in budding yeast, we show that two major cell cycle kinases, Dbf4-dependent Cdc7 kinase (DDK) and Polo-kinase (Cdc5), collaborate to link the mode change of HR to the meiotic cell cycle by. This regulation of HR is through the SC. During prophase I, DDK is shown to maintain SC integrity and thus inhibition of Rad51. Cdc5, which is produced during the prophase I/metaphase I transition, interacts with DDK to cooperatively destroy the SC and remove Rad51 inhibition. By enhancing the interaction between DDK and Cdc5 or depleting DDK at late prophase I, meiotic DNA breaks are repaired even in the absence of Dmc1 by utilising Rad51. We propose that the interplay between DDK and Polo-kinase reactivates mitotic HR mechanisms to ensure complete repair of DNA breaks before meiotic chromosomem segregation.

Endogenous nitric oxide (NO) produced by the enzyme NO synthase (NOS) has an important role in a variety of physiological processes. However, NO becomes noxious to cells if produced in excess. Therefore, the production of NO is tightly regulated. A particularly exciting and novel aspect of the regulation of NO signalling is the possibility that the expression of NOS genes is controlled by unconventional mechanisms that depend on the presence of natural antisense transcripts (NATs). In this thesis I investigate the important properties of two distinct NOS-related NATs: Lym-antiNOS2 and Mm-antiNOS1. I show that Lym-antiNOS2 RNA is widely expressed in the CNS of the pond snail Lymnaea stagnalis. Furthermore, I demonstrate that the expression of Lym-antiNOS2 is differentially regulated by training leading to long-term memory formation. Moreover, my results indicate that Lym-antiNOS2 RNA is subjected to peripheral trafficking in neurons. As for Mm-antiNOS1, I find that its expression is restricted to embryonic brain tissue and is almost undetectable in the adult brain of Mus musculus.

In response to DNA double stranded breaks (DSBs), mammalian cells have evolved two major repair pathways, DNA Non Homologous End Joining (NHEJ) and Homologous Recombination (HR). The majority of DSB repair in G1 and G2 phase is repaired with fast kinetics by NHEJ in a pathway that involves the core NHEJ factors: Ku, DNA-PKcs, XLF, DNA Ligase IV and XRCC4. A subset of slow repairing DSBs also requires ATM and Artemis (Riballo et al, 2004). This slow component of repair represents DSBs that reside within highly compacted regions of the genome known as heterochromatin (HC) (Goodarzi et al, 2008). ATM functions at HC to mediate relaxation by phosphorylating the HC building factor KAP-1 (Goodarzi et al, 2008). Here I provide evidence that DSBs dependent upon Artemis for their repair also reside within regions of HC. However, unlike ATM, Artemis functions downstream of the HC relaxation process. In response to DSBs, ATM phosphorylates the histone variant H2AX (γH2AX). γH2AX acts as a docking site for the localized recruitment and activation of DNA Damage Response (DDR) proteins. The expansion of γH2AX can spread over megabases of DNA. Here I have shown that highly compacted KAP-1, MeCP2 and DNMT3B enriched chromatin acts as a barrier to IR induced γH2AX expansion. In patient cells deficient for MeCP2 or DNMT3B proteins, such as Rett syndrome (MeCP2 deficient) and Immunodeficiency centromeric-instability facial-anomalies syndrome (DNMT3B deficient), ATM and Chk2 signalling is heightened, which is reflected in a hypersensitive and prolonged G2/M checkpoint arrest. These findings suggest that higher order chromatin complexity is a barrier to ATM signalling to the checkpoint machinery. In the final section of my thesis, I addressed what affect DNA damage complexity exerts on checkpoint arrest. Using exposure to heavy ion irradiation, which induces complex DSBs, I observed larger γH2AX foci and prolonged G2/M checkpoint arrest.

Understanding the mechanisms and factors that govern cell cycle control is key to developing more effective treatments for human diseases, such as cancer. The human MASTL gene encodes an AGC family kinase, Greatwall kinase, that is conserved among higher eukaryotes. The protein contains an unusual bifurcated kinase domain separated by a stretch of nonconserved amino acids. In Drosophila, mutations in Greatwall cause the failure of chromosomes to condense resulting in a delayed entry in to and progression through mitosis. In mitotic Xenopus egg extracts, immunodepletion of Greatwall results in exit from M phase, characterised by the decondensation of the chromosomes and the reforming of the nuclear envelope. The addition of purified Greatwall to egg extracts immunodepleted for Greatwall causes precocious phosphorylation of Cdc25 and premature entry into mitosis. These reports indicate that Greatwall plays an important role in the control of mitosis but little is known about the function of Greatwall kinase in human cells, its structure, or control of its activity. This project aimed to elucidate the role of this novel kinase in human cells. To this end, the gene has been cloned and antibodies generated to allow the study of human Greatwall kinase. RNAi-mediated knockdown of Greatwall in HeLa cells caused aberrant mitotic progression and apoptosis. To gain further insight into the mechanism of Greatwall activation, the Greatwall kinase structure was modelled and key motifs of the kinase fold identified. In particular, a key activating phosphorylation was identified, and a specific antibody raised to this site, allowing investigation of the regulation of Greatwall activity at mitotic entry and exit. The use of chemical genetics to attempt to specifically inhibit the kinase in human cells is described. Finally, evidence is presented that Greatwall kinase may represent a promising new biomarker and drug target for cancer therapy.