Search

Synthetic glucocorticoids (Gc) are anti-inflammatory agents, which are commonly used in cancer treatment. Glioblastoma (GBM) is an aggressive form of brain cancer with a median survival of 15 months. Dexamethasone, a potent Gc, is used to reduce oedema, however, patients prescribed high doses of Dexamethasone have reduced survival times. Despite the widespread use of Gc, much of their mechanism of action is still unclear, and their effects are tissue-specific. This project has therefore aimed to investigate Gc function in the context of GBM. Gc activate the glucocorticoid receptor (GR), a ligand activated transcription factor. Through RNA sequencing, Gc were predicted to affect DNA repair pathways in GBM, through control of p53 effectors. We demonstrated that Gc treatment increased DNA repair within GBM cells. This is relevant in the context of GBM treatment, as radiotherapy and chemotherapy both rely on the induction of DNA damage to induce GBM cell death. Screening selective steroids which fail to induce these transcriptional targets shows promise in maintaining beneficial anti-inflammatory effects while maintaining efficacy of standard treatments. Evidence is emerging that tumour cell populations show plasticity and can therefore adapt their response over time, and evade cancer treatment. Understanding the underlying mechanisms that control Gc responses at the single cell level is therefore essential to understand Gc action in vivo. Single cell RNA sequencing revealed that even within a clonal cell population, gene expression was variable. While the expression of GR correlated with the expression of GR target genes, there was less correlation between GR target genes suggesting importance of other unidentified factors. This work uncovered a new mechanism of GR action through DNA repair, and revealed for the first time how Gc responses are controlled at a single cell level. These insights will lead to improved treatment strategies for GBM and allow for the further study of factors which modulate the Gc response in individual cells.

Primary cilia are microtubule-based “antennae-like” organelles extending from the apical surface of most mammalian cells. They integrate mechanical or chemical signals essential for cell homeostasis and differentiation through several cooperating compartments. Mutations in genes that encode ciliary proteins or proteins essential for correct ciliary function are the cause of a major group of inherited and variable developmental disorders known as ciliopathies. Functional interactions between ciliary compartments, the molecular basis of variable clinical phenotypes, and the mechanisms of cilia formation are all still poorly understood. The presence or absence of cilia can be easily imaged and quantitated. This lends itself to high-throughput, high-content imaging in reverse genetic screens of cellular phenotypes. siRNA reverse genetic screens were performed to assess increased cilia incidence and identified a key negative regulator of ciliogenesis. Analysis showed this regulator acts through a mechanism involving actin remodelling and acto-myosin contraction. Pharmacological inhibition of this target may therefore comprise a novel therapeutic approach for treatment of a broad group of ciliopathy disease classes. Further screening of a data-set for supernumerary primary cilia, added to the evidence that cytokinesis is not a prerequisite for ciliogenesis. Disruption of the centralspindilin complex caused mitotic failure and maturation of supernumerary centrosomes, leading to the formation of the supernumerary cilia, a known cellular phenotype of severe ciliopathies. Finally, a combinatorial screening approach to generate double genetic perturbations of ciliary genes identified reciprocal synthetic genetic interactions between anterograde intraflagellar transport (the IFT B complex) and the transition zone. Reverse genetics screening techniques have identified novel regulators and pathways of ciliogenesis, and a potential therapeutic target for ciliopathies. Furthermore, combinatorial screening has highlighted a novel and complex interaction in ciliary biology, that may provide potential new insights into ciliary organisation and disease pathomechanisms.

Autosomal recessive primary microcephaly (MCPH) is a disease consisting of a small brain and associated intellectual disability. The most common cause of MCPH is mutation of the abnormal spindle-like microtubule-associated gene (ASPM, formerly known as abnormal spindle-like microcephaly-associated gene) at the MCPH5 locus. This thesis studies the function of ASPM, mitotic alterations due to homozygous ASPM mutation and identifies ASPM C-terminal interactants. ASPM siRNA KD studies confirmed previously identified roles for ASPM in spindle pole focusing, spindle/cleavage furrow orientation, spindle assembly and cytokinesis. Roles for ASPM in maintaining nuclear and centrosome structure, centrosome biogenesis, mitotic progression, microtubule organisation, and abscission were identified as being microtubule associated. A novel role for ASPM in actin organisation was identified. In MCPH5 patient fibroblast cells carrying the homozygous nonsense ASPM mutation 3663delG, a low level of full length ASPM protein was observed suggesting transcriptional read-through was occurring. ASPM3663delG was mis-localised to the cytoplasm suggesting it has a reduced ability to either travel to the mitotic spindle pole or become associated at the pericentrosomal matrix. In the patient fibroblast cells carrying ASPM3663delG and ASPM9984+1T > G homozygous mutations microtubule defects caused a variety of aberrant mitotic phenotypes which induced an increase in the duration of mitosis. These defects included prometaphase (ASPM3663delG) or metaphase (ASPM9984+1G > T) arrest, spindle mis-orientation and abscission failure and a decrease in the incidence of proliferation. Therefore MCPH arises III from a combination of mitotic abnormalities rather than a single aberration, which together result in a reduction in neural progenitor cell number. It is proposed that the single MCPH5 small brain phenotype arising in patients carrying homozygous nonsense ASPM mutations may be the result of transcriptional read-through. ASPM C--terminus has been shown to be required for ASPM spindle pole localisation, microtubule organisation and cytokinesis. Microtubule actin cross-linking factor 1 (MACF1) was identified as a candidate ASPM interacting protein from a yeast-2 Hybrid screen of the ASPM C-terminus against a foetal brain cDNA library. The MACF1-ASPM association was validated by co-immunoprecipitation (co-IP). MACF1 was shown to be associated with microtubules during interphase and was dependent upon ASPM for this localisation. MACF1 mitotic localisation was consistent with ASPM localisation at the minus ends of spindle microtubules and pericentrosomal matrix from prometaphase to early telophase. In patient cells carrying an ASPM C-terminal mutation MACF1 protein levels were proportional to ASPM expression. The alteration in the ASPM-MACF1 association due to ASPM mutation or ASPM reduction may account for the changes in actin organisation, microtubule associated forces, spindle mis-orientation and abscission observed in this research study.

Familial exudative vitreoretinopathy (FEVR) is a genetically heterogeneous blinding disorder characterised by the abnormal development of the retinal vasculature. Genetic studies have led to the identification of eight different genes mutated in FEVR and an additional autosomal dominant locus (EVR3) has been mapped. Recently the EVR3 locus was sequenced using next generation technology leading to the identification of an LGR4 missense variant in the EVR3 family. Subsequent sequencing of LGR4 in a cohort of FEVR patients identified further missense variants. The work in this thesis describes the functional characterisation of LGR4 and its FEVR-related variants to confirm it as a new FEVR disease gene. Zebrafish lgr4 morpholino knockdown followed by mRNA rescue showed that the FEVR-related LGR4 variants were unable to rescue the retinal vasculature defects induced in the fish. Furthermore, the majority of mutations underlying FEVR encode components of the Norrin-ß-Catenin signalling pathway. The TOPflash ß-Catenin reporter assay was used to show that LGR4 potentiates Norrin signalling but the variants located in the binding domain of LGR4 reduce this. Norrin and LGR4 binding assays show that the EVR3 mutation increases the binding affinity between these two proteins hinting at a potential disease mechanism. Finally, an in vitro angiogenesis assay demonstrated that LGR4 plays a role in the development of vascular structures. The identification of LGR4 as a new FEVR gene, and confirmation that LGR4 is a component of the Norrin-ß-Catenin pathway, helps to decipher the molecular mechanisms underlying the normal development of the retinal vasculature and the FEVR disease mechanism. Knowing the gene underlying a disease translates into immediate benefit for patients and families through access to a precise genetic diagnosis and more accurate genetic counselling. Furthermore, this new understanding should contribute to the development of new treatments or therapy, ultimately providing a better quality of life for the patients.

Parkinson’s disease (PD) is characterised by loss of dopaminergic neurons within the substantia nigra pars compacta. Mitochondrial dysfunction and oxidative stress are pathological characteristics of PD. It is not fully understood how neurons respond to these stresses, though adaptive responses involving the activity of PD-associated proteins such as Parkin may play a critical role. When these adaptive responses are overcome neurons are lost through apoptotic mechanisms aimed at minimising further tissue damage. AP-1 transcription factors are encoded by immediate early genes (IEGs) such as FOS and JUN. They regulate early cellular responses to stress. Mitogen-associated protein kinases (MAPKs), such as JNK and ERK, regulate AP-1 function. Previous studies have implicated MAPK/AP-1 signalling in the progression of PD. The aim of this thesis was to elucidate whether the AP-1 system was involved in the response to mitochondrial stress in neurons and to dissect the signalling pathways by which this was regulated. SH-SY5Y neuroblastoma cells were used as a model system. CCCP-induced mitochondrial uncoupling induced oxidative stress in this cell line. In response to this AP-1 levels were modulated at both the mRNA and protein levels. SH-SY5Y differentiation or Parkin overexpression altered AP-1 protein response profiles. Treatment with c-Jun siRNA lead to an increase in cell death induced by lower levels of mitochondrial coupling, but a decrease under higher levels of uncoupling. Additionally, a novel link between Parkin and the AP-1 response was observed and investigated. Further work indicated that both JNK and ERK could regulate c-Jun function. Critically, this was dependent on both cell differentiation and the level of stress. Notably, JNK signalling was the primary promoter of apoptosis in response to high levels of mitochondrial stress in differentiated cells. This study suggests that the JNK/c-Jun pathway is differentially regulated by the level of mitochondrial stress, contributing to both adaptive and apoptotic responses and thereby playing a central role in deciding neuronal fate. It further suggests that enhancing adaptive cytoprotective pathways represent a better approach than inhibiting apoptotic pathways when developing therapeutics for the neurodegenerative progression observed in PD.

Mutations in the N-terminal region of MCPH1/BRIT1 cause premature chromosome condensation (PCC), whereby cells enter mitosis before completing DNA replication. 792 chemical compounds (CC) were selected based on the crystal structure of the N-terminus of MCPH1/BRIT1 and assayed using high throughput- high content imaging to identify CC that induced PCC. Hit validation revealed 4 potential CC, 2 of which induced high PCC at low concentrations. A screen using a human protein kinase (hPK) siRNA sub-library was performed to identify genes that induced PCC. Four hits were selected for validation, however PCC induction was not confirmed. A complementary hPK siRNA screen combined with MCPH1/BRIT1 siRNA knockdown was performed. The cell number outputs from both hPK siRNA screens were analysed to identify synthetic lethal (SL) genes in MCPH1/BRIT1-deficient cells. CDK1/CDC2, STK39, VRK1 and TTK/MPS1 were subsequently validated as potential MCPH1/BRIT1 SL genes. The expression of MCPH1/BRIT1 was examined by immunostaining in breast cancer (BC) tissue pre and post neoadjuvant chemotherapy (NACT) to determine its effect on response and survival. MCPH1/BRIT1 expression increased in response to NACT with high expression in 51.4% (36/70) of cases pre-NACT compared to 81.4% (57/70) post-NACT (p = 0.0002). Reduced MCPH1/BRIT1 expression correlated with longer overall survival (OS) pre- but not post-NACT (p = 0.017). Change in MCPH1/BRIT1 expression (from low-high) post-NACT was significantly correlated with better OS (p = 0.010). MCPH1/BRIT1 has previously been found to regulate p53 stability in BC cell lines. Notably, in this study a significant increase in MCPH1/BRIT1 staining was accompanied by a decrease in p53 staining in post-NACT samples (p < 0.0001). In conclusion, these data support the idea that CC inhibitors targeting MCPH1/BRIT1 may sensitize BC cells to chemotherapy. Additionally, genes whose inhibition could promote cell death in MCPH1/BRIT1–deficient cells have been identified as potential therapeutic targets in tumours where MCPH1/BRIT1 expression or function has been compromised.