Search

Over 38.4 million people worldwide are living with HIV-1, the etiological agent of acquired immunodeficiency syndrome and the cause of over 40.1 million deaths. The latent HIV-1 reservoir is the major roadblock to cure and necessitates lifelong antiretroviral therapy (ART), which is vulnerable to drug resistance. I contributed towards the development of a novel therapeutic strategy that aims to eliminate cells actively expressing HIV-1 and those harbouring HIV-1 reactivated from latency. This approach hijacks the currently unexploited HIV-1 alternative RNA splicing process to functionalize a defective cell suicide enzyme (*HSVtk*ΔAUG1) through targeted RNA *trans*-splicing with the HIV-1 D4 splice site. In-frame *HSVtk* translation is initiated from the start codon of the HIV-1 *tat1* donor exon in chimeric mRNA. HSVtk activates the prodrug ganciclovir (GCV), with cytotoxic metabolite GCV-triphosphate (TP) disrupting DNA synthesis to selectively kill HIV-1-expressing cells. Following proof-of-principle transfection studies, therapeutic constructs were engineered for lentivirus-mediated delivery to HIV-1-infected cells in vitro. I first optimized production of a panel of VSV-G-pseudotyped D4-targeting lentivectors for high infectious titre and low transfer plasmid carryover. I confirmed trans-splicing between HIV-1 *tat1* and *HSVtk*ΔAUG1 RNA in Jurkat T cells co-transduced with HIV-1NL4-3ΔE and therapeutic vectors. However, I found that translation of catalytically active polypeptides from internal AUGs in *HSVtk*ΔAUG1 caused dose-dependent cytotoxicity with GCV in uninfected cells. Modification of internal AUGs in D4 opt 2 effectively mitigated this major off-target effect. Based on the MTT assay, D4 opt 2 in combination with GCV reduced the viability of HIV-1NL4-3ΔE-expressing Jurkat T cells by approximately 50% with no impact on uninfected cells. For improved therapeutic potential, I replaced the promoter in D4 opt 2 with that for the human *EF1α* gene, resulting in 4.4-times more RNA payload. EF1α-driven D4 opt 2 with GCV reduced the viability of HEK293T cells expressing full-length HIV-1NL4-3 by up to 70%. I developed a panel of lentivectors delivering opt 2 with low (ScrambleV2), moderate (ScrambleV1), or high (D4) affinity predicted in silico for target HIV-1 pre-mRNA and found that the strength of interaction between opt 2 and HIV-1 correlated with the propensity for *trans*-splicing and elimination of HIV-1-expressing cells by the opt 2 cell suicide system *in vitro*. Having demonstrated that this HIV-1-targeted cell suicide system could eliminate cells actively expressing HIV-1, I next investigated its potential against latent HIV-1 in J- Lat 10.6 cells. I found LRAs exerted class and dose-dependent effects on viability which could affect the outcome of shock and kill. Despite their superior reactivation potential, NF-κB agonists made J-Lat 10.6 cells more difficult to eliminate. In contrast, the 26S proteasome inhibitor bortezomib (BTZ), known to sensitise HIV-1-infected cells to death, enhanced shock and kill. I discovered that the nucleoside analogue DNA methyltransferase inhibitor decitabine (DAC) is an adjuvant of the HIV-1-targeted cell suicide system, with DAC-TP and GCV-TP known to synergise. A ≥65% reduction in J-Lat 10.6 viability was achieved when cells were stimulated with BTZ, DAC, or a TNFα/DAC 'shocktail' prior to treatment with EF1α-driven D4 opt 2 and GCV. My work supports further development of our HIV-1-targeted cell suicide system against cells actively expressing HIV-1 and those chronically infected when combined with appropriate LRAs.

SYNGAP1 (Synaptic Ras GTPase activating protein 1) haploinsufficiency-associated disorder or intellectual developmental disorder, autosomal dominant 5 (MRD5) is caused by autosomal dominant loss-of-function mutations in the SYNGAP1 gene. De novo SYNGAP1 mutations represent one of the most common causes of non-syndromic intellectual disability (NSID) (OMIM #603384). The most common symptoms include moderate-to-severe intellectual disability (ID), epilepsy and autism spectrum phenotypes. SYNGAP1 protein is a component of the postsynaptic density complex where it acts as one of the key effectors of N-methyl-D-aspartate (NMDA) receptor downstream pathways. It is involved in regulatory pathways such as the trafficking of the α-amino-3-hydroxy-5-mehyl-4-isoxazolepropionic acid (AMPA) receptors, regulation of AMPA receptor translation and the regulation of several forms of synaptic plasticity. With the aim to study SYNGAP1 protein function and its involvement in the pathophysiology of the disease, several rodent models have been created and extensively characterised. These studies revealed phenotypes which partially resemble the symptomatology observed in humans such as seizure susceptibility, hyperactivity and impulsivity. It has been shown that early restoration of normal SYNGAP1 protein levels prevents the onset of haploinsufficiency-associated phenotypes in the mouse. This, and the monogenic origin of the disorder, makes SYNGAP1 a candidate target for gene replacement therapeutic approaches. I hypothesised that restoration of SYNGAP1 levels via Adeno Associated Virus (AAV)-mediated gene transfer during early postnatal development, will prevent the appearance or ameliorate the severity of behavioural phenotypes observed in the mouse model of SYNGAP1 haploinsufficiency. Therefore, the aim of this project was to evaluate molecular therapies to test this hypothesis. To this end, I first defined a robust battery of behavioural tests in order to confirm and identify deficits to use as endpoints to evaluate the efficacy of gene therapy products. Well-established and novel phenotypes were evaluated in wild-type and Syngap1+/- mice at different postnatal developmental time points. Syngap1+/- mice showed increased locomotor activity, anxiety-like phenotype and a risk-taking like behaviour. No genotype effect was observed in tests assessing compulsive behaviour, working memory, motor function and general well-being and depth perception. For the first generation of candidate therapeutic constructs two different promoters were used, the minimal murine Mecp2 promoter (MeP) 229 and the synthetic JeT promoter. MeP229 regulated constructs were expressing three different Myc-tagged human SYNGAP1 isoforms, A2, A1 and B2 while the JeT regulated construct the Myc-tagged human SYNGAP1 isoform A1. In vitro evaluation of protein expression revealed low levels of plasmid-derived SYNGAP1 produced from the MeP229 regulated constructs while a higher amount of Myc-tagged SYNGAP1 protein was detectable from the JeT regulated cassette via both fluorescence immunocytochemistry and immunoblotting. Protein expression from the vectorised ssAAV9/JeT-hSYNGAP1_A1-Myc was confirmed in the brain in vivo at 5 weeks post intracerebroventricular injection. A therapeutic dose escalation efficacy study assessing ssAAV9/JeT-hSYNGAP1_A1-Myc was then conducted in Syngap1+/- mice (1E10, 5E10 and 1E11 vg/mouse). Overall, these data revealed no amelioration of hyperactivity or anxiety phenotypes, but a modest trend toward the amelioration of the risk-taking behaviour was observed in a platform departure test. Post-hoc immunoblot analysis confirmed a dose-dependent expression of the vector-derived SYNGAP1. However absolute levels of SYNGAP1 protein measured in different brain areas were below physiological levels of the native protein. Subcellular localisation analysis in whole hippocampal synaptosomal preparation showed that the vector-derived protein was correctly translocated to the synaptic compartment, mimicking endogenous patterns of protein localisation.

Introduction: Fragile X Syndrome (FXS) is an X-chromosome linked, neurodevelopmental disorder, which is caused by the silencing of the Fragile X Messenger Ribonucleoprotein 1 (FMR1) gene that encodes the Fragile X Messenger Ribonucleoprotein (FMRP) protein. FMRP is ubiquitously expressed and is especially abundant in the brain and testes. There are multiple isoforms of FMR1 but the precise function of the protein remains to be fully characterized. FXS is the most common form of inherited intellectual disability and the leading single gene cause of autism, affecting approximately 1 in 4000 males and 1 in 8000 females worldwide. FXS is considered a good candidate for gene therapy as it is a monogenic disorder with a severe manifestation and has no disease modifying treatment options. Aim: The aim of this project was to test novel AAV9/FMR1 gene therapy vectors both in-vitro and in-vivo, to establish efficacy and safety of these vectors and to assess if re-expression of FMRP can prevent the development of FXS-associated phenotypes in a mouse model of FXS. Methods: Three novel single stranded AAV9/FMR1 (FXS4, FXS5 & FXS6) gene therapy vectors (designed by the Cobb lab) underwent quality control assessments and were evaluated in two in-vitro models, Neuro-2a cells and HEK-293 cells prior to testing in a mouse model of FXS. The most promising vector (FXS6) was identified and used for larger therapeutic prevention trials. This promising novel cassette delivers a dominant human FMR1 isoform, isoform 7, and includes native elements of the human FMR1 promoter and 3'UTR sequence. Neonatal wild-type (WT) and Fmr1-/y mice were dosed with either the FMR1 gene therapy vector or excipient control via intracerebroventricular injections. At two months post injection, mice were evaluated using a battery of behavioural tests. In addition, mice were assessed for seizure threshold and severity using an auditory stimulus protocol. Following the in-life phase, mouse brains were assessed for vector biodistribution, transduction efficiency and vector derived FMRP protein levels. Results: All three vectors passed the quality control tests and proceeded with virus production. In-vitro testing revealed transcript being produced during Neuro-2a cell transfections using all three constructs. However, protein analysis revealed limited vector-derived protein expression and low transduction efficiency with all but one construct, FXS4. Initial pilot in-vivo testing using FXS4 resulted in surprisingly low transduction efficiency. Nevertheless, following further pilot in-vivo testing of all three vectors, only one was identified as promising, FXS6, indicated by high cell transduction levels. The initial assessment trial of the lead ssAAV9/FMR1 vector (FXS6), achieved relatively moderate and variable transduction efficiency ranging between 20-60% of positive nuclei in the cortex, hippocampus, thalamus, and inferior colliculus expressing vector derived FMRP. Fmr1-/y mice showed a profound genotype effect when assessed for audiogenic seizures with high a penetrance of seizures (>80% of subjects) in vehicle treated Fmr1-/y mice relative to WT littermates (10-25%). Cohorts treated with lead ssAAV9/FMR1 vector showed no improvement in seizure incidence or severity. No statistically significant improvement was observed in the marble burying test and open field test. An important caveat of this trial was the mouse model used, which had a mutation in a gene causing retinal degeneration and thus questioning the behavioral data obtained in this study. Additionally, the dose injected (5*1011 vg/mouse) and the resulting transduction efficiency didn't correlate with the literature, putting into question the quality of the virus. To address these issues, a secondary therapeutic prevention trial was completed, using a sighted Fmr1-/y mouse model and new virus manufactured by a separate vendor. The second therapeutic prevention trial using FXS6, assessed mice dosed at 5 different doses of the vector ranging from 5*109 -3*1011 vg/mouse. The gene therapy vector was well tolerated in wild-type mice and did not produce any overt adverse neurological observations at any of the doses tested. Cohorts treated with the vector showed a strong and dose dependent reduction in the occurrence and severity of induced seizures with the higher doses rescuing the induced seizure burden to WT levels. Behavioral analysis revealed improvements in marble burying for Fmr1-/y mice treated with the two highest doses but no clear treatment effect in the open field or elevated plus maze tests. Ex-vivo analysis of brain samples from treated mice revealed a dose dependent increase in the brain transduction efficiencies (2-20% with the lowest dose and 68-92% with the highest dose, dependent on the brain region). Vector derived FMRP expression, validated by western blot, also varied depending on the brain regions, and was measure at between approximately 5% and 200% of the WT level with the lowest dose and the highest dose respectively. Conclusion: The work done throughout this project illuminates the potentials of gene therapy as a promising therapeutic avenue for complex monogenic disorders that have no effective treatment. The main findings of the project demonstrate the effectiveness and safety of an AAV9/FMR1 gene therapy vector utilizing native regulatory elements in ameliorating disease-related phenotypes in mice modeling FXS. Future studies aim to assess neurological and behavioral outcomes in the more robust rat model of the disorder as well as utilize new and emerging technologies in the field of gene therapy.

The use of Lentiviral Vectors (LVV) as tools for gene delivery has gained momentum in recent years due to successes in clinical trials for rare diseases. Production of high-titre LVVs is a limiting factor in their wider application in diseases with large patient populations, such as cancer. LVVs are commonly produced by the transient transfection of HEK293T cells, however this process is costly and inefficient. An alternative stable, scalable manufacturing system for LVV production is necessary. Before the development of a stable cell line can be achieved, it is vital to gain an understanding of how individual vector components interact with HEK293T cells at a molecular level. HIV-1 GagPol is the major viral structural protein constituting approximately 70% of a bona fide virion. It is not known how synthetic GagPol, utilised during LVV production at GSK, interacts with HEK293T cells. Given that HIV-1 utilises several host cell systems for its own propagation, it is hypothesised that synthetic GagPol may have similar effects, potentially stressing the capacity of a HEK293T cell to carry out typical homeostatic functions. This could be detrimental to HEK293T cells in terms of causing cellular stress due to the burden of ectopic protein production, and may also impact negatively on GagPol production. This thesis therefore aimed to develop an understanding of the interactions between synthetic GagPol and HEK293T cellular physiology in order to define approaches to identify and intervene in molecular bottlenecks associated with LVV production. To do this, plasmid constructs encoding GagPol and fluorescently-tagged GagPol were generated and used in HEK293T transient transfections, where a maximum GagPol production titre of over 1250 ng p24/mL was achieved. HEK293T cellular physiology was examined during high-level GagPol production. Activation of the unfolded protein response (UPR) and endoplasmic reticulum (ER) stress pathway, in addition to the induction of apoptosis and autophagy were observed. The effects of chemical inhibitors of cellular stresses were examined, with chemical inhibitors of apoptosis leading to a modest increase in GagPol production, in addition to increased cell viability. These investigations with chemical inhibitors of cellular stress illuminate the potential for genetic targets of manipulation within HEK293T cells that could enable increased GagPol production which would, in turn, be beneficial for LVV production.

Gene delivery will cure all. That was the hope of many since the discovery and therapeutic implementation of genetic research. Although some of that hope has been realised, the delivery of nucleotides to cells in the body remains a major barrier in wider application. Ultrasound-mediated delivery of drugs has been investigated in various forms for decades and continues to be an area of engineering and science that is much explored. The body of work presented in this thesis strides to bring the two areas of research endeavour together. Ultrasound-mediated delivery of drugs and antibody particles, normally in conjunction with cavitation nucleation agents, has been shown to be effective both in vitro and in vivo. In this work, we focus on nucleotide delivery, in particular mRNA, which has yet to be fully explored, especially in the context of achieving delivery of non-encapsulated, 'free' mRNA. Data presented here explores the relationship between cavitation and the transfer of, and protein expression from, free mRNA. The protein expression exhibited in various cell lines in vitro was further tested in vivo, using either sub-micron or micron-sized cavitation nuclei and the effects on transfection efficiency and expression time were examined. A luciferase reporter gene assay was used to characterise events in vitro across different cell lines including cancer and non-cancerous types. The impact and varying acoustic exposure parameters and cavitation on expression were assessed, and differences in expression time and transfection rate were observed. These data were used to determine appropriate acoustic parameters for in vivo delivery. Cavitation-mediated delivery was examined in vivo in two types of cancer tumours, and mRNA expression was quantified using in vivo fluorescence imaging and using excised homogenised tumour tissue. The results presented here lay the foundations for cavitation-mediated delivery and transfection of free mRNA both in vitro and in vivo by identifying appropriate acoustic parameters and cavitation nucleation agents, which result in enhanced delivery and by identifying the appropriate time windows for measurements to quantify expression.

All somatic cells in the human body originate from the fertilised egg, or 'zygote'. Co-ordinated processes of cell division and differentiation result in the incredible complexity observed following development. Thereafter, each tissue must maintain its function for the decades of life that follow, while minimising risks such as cancer. Crucial to understanding the strategies employed to meet these demands is knowing what cell gives rise to what i.e. the structure of the somatic lineage tree. For this purpose, it is fortuitous that most cell divisions result in the acquisition of somatic mutations that are passed on to all of its future progeny. Therefore theoretically, if all somatic mutations in each cell of an organism were known, the organism's full somatic lineage tree could be determined. While this is not yet possible at the scale of an entire organism, advances in clonal expansion and whole-genome sequencing technologies have facilitated such experiments on the scale of hundreds of cells in a single individual. This thesis contains three results chapters, each utilising somatic phylogenies in different ways. The first looks at early human development. Due to the small number of cells in early development, a near complete phylogeny can be determined. Assessing contributions of early lineages to different tissues is used to determine the developmental relationships and origins of tissues. The second looks at the clinically important context of gene therapy for sickle cell disease. Comparisons of somatic phylogenies from before and after the gene therapy procedure gives insights into the biology of sickle cell disease and the safety of gene therapy in this context. Finally, the third chapter looks at somatic phylogenies from a different perspective: as a way of testing our assumptions regarding mechanisms of mutation acquisition. The observation that some mutations do not fit these assumptions leads to the surprising conclusion that some mutation-causing DNA lesions persist unrepaired for months to years. Overall, this thesis demonstrates the power of interrogating somatic phylogenies to answer important questions in diverse biological disciplines. As technological advances allow such experiments on ever larger scales, the power of this approach will only increase.

Adult stem cells become less efficient at maintaining tissue integrity and repairing tissue damage with age. Oligodendrocyte progenitor cells (OPCs) are the largest population of stem cells in the adult CNS and give rise to oligodendrocytes. OPCs play a critical role in myelin regeneration (remyelination), but, like all stem cells, their regenerative capacity declines with age. OPCs progressively fail to differentiate into mature oligodendrocytes following a demyelinating insult with age, and this failure plays a key role in the progression of prominent neurodegenerative and demyelinating diseases, such as Alzheimer's disease and multiple sclerosis. I have optimised an existing system for isolating OPCs from aged rats to assess rejuvenation strategies in vitro, designed novel OPC-specific rAAV vectors to manipulate gene expression in these cells, and used these tools to show that overexpression of MYBL2 in aged OPCs increases their differentiation capacity in vitro. My work adds to a growing body of literature showing that overexpression of specific factors in specific tissues is sufficient to rejuvenate their resident aged cells.

Haematopoietic stem cell (HSC) ex vivo gene therapy is now successfully used to treat an in-creasing number of monogenic disorders affecting the blood system. The ultimate efficacy of this therapy depends upon successful targeting of long-term haematopoietic stem cells (LT-HSCs), which are characterised by a predominantly quiescent status and sustained self-renewal capacity. Clinical ex vivo gene therapy protocols target the heterogenous mix of stem and pro-genitor cells encompassed in the CD34+ fraction with a limited understanding of how ex vivo manipulation impacts a purified LT-HSC subset. Importantly, long-term repopulation capacity is lost over culture, although the kinetics and molecular drivers of this process remain unclear. To address these questions, I perform single cell RNA-Seq, in vitro functional assays and in vivo transplantation in a time resolved manner following cord blood (CB) culture in differentiation facilitating conditions and mobilized peripheral blood (mPB) culture during a lentiviral ex vivo gene therapy protocol. First, I characterise the molecular impact of a 62 hr lentiviral ex vivo gene therapy protocol on mPB LT-HSC, short-term HSC (ST-HSC) and CD34+ cells. I reveal that the ex vivo protocol dramatically rewires the quiescent LT-HSC transcriptome and demonstrate that the majority of transcriptional changes are attributed to the culture conditions rather than lentiviral transduction. Secondly, I refine the kinetics associated with HSC functional attrition over the first complete cell cycle ex vivo. Long-term repopulation capacity is maintained for the first 6 hr irrespective of HSC source or tested culture conditions. I identify the 6 hr time-point as encompassing an adaptation period where LT-HSCs are rapidly responding to the instructive signals of culture and is molecularly underpinned by transient upregulation of cell stress response signalling. Follow-ing the 6 hr time-point, long-term repopulation capacity drops dramatically by 24 hr in CB culture and 62 hr in mPB culture. Loss of HSC function is correlated with reduced survival, cell cycle progression (late G1-M), sharp upregulation of MYC and a reduced ability to resolve proteostatic stress. Taken together, my results demonstrate that the 6 to 24 hr transition point is instrumental for the determination of LT-HSC cell fate ex vivo. Finally, using an ex vivo culture system of reversible early G1 arrest, I formally test whether cell cycle progression drives loss of self-renewal capacity in culture. Long-term in vivo transplantation approaches conclusively establish that cell cycle progression is not responsible for the loss of long-term repopulation capacity associated with clinically relevant HSC culture.

Telomerase hTERT RNA is overexpressed in around 90% of all cancers but targeting it has been unsuccessful to date due to the inability of this approach to kill telomerase-expressing cells leading to the evolution of telomerase-independent cells. The approach proposed in the thesis (NASPER), aims to target cells overexpressing hTERT RNA and cause their apoptosis, preventing this evolution and debulking the tumour mass. NASPER involves bringing two fusion proteins, each of which comprises a custom-designed PUF (Pumilio and FBF) RNA-binding protein and a protease, onto the hTERT RNA into close proximity to activate the protease which should lead to cell death. The proteases tested in this study are the HIV protease (HIVPR) and the split-TEV protease, which require dimerisation/fragment complementation for catalytic activity. In Chapter 3, the two designed PUF proteins were first purified and tested in vitro using fluorescence polarisation experiments to assess RNA binding. The results indicate that the PUF proteins bind specifically to their cognate sequences both independently and in combination. In Chapter 4, expression in E. coli and purification of HIVPR was optimised, and its activity was confirmed in vitro. In cells, the individual PUF-HIVPR fusion proteins appeared to auto-activate without their dimerisation partner despite the use of mutations to reduce auto-activity, and it was concluded that a split system is required such that the protease can only undergo activation when both proteins are present. In Chapter 6, a split-GFP system (fusing two GFP fragments to each of the two PUF proteins) most notably show that, as intended, the fusion proteins bind better when cognate RNA is used versus scrambled RNA, and that engineering the fusion proteins such that the split-GFP are domains oriented towards each other results in improved re-constitution of GFP. Subsequently, a NASPER system designed with the split-TEV protease was tested, in which two TEV protease fragments are fused to each of the two PUF domains. The results were consistent with the split-GFP findings, and NASPER was also able to target overexpressed hTERT mRNA in HeLa cells. No cleavage of a procaspase-3 construct could be detected in apoptosis assays in Chapter 8, and further work is required to elucidate the cause. Chapter 9 describes further experiments to understand the binding characteristics of designed PUF domains using crosslinked RNA-seq methods. Overall, the results presented in the thesis provide new insights into the behaviour of designed PUF proteins in the cell and lay the groundwork for a new therapeutic approach based on targeted protease-induced cell death, as discussed in Chapter 10.

The aim of this thesis was to evaluate the efficacy and safety of retinal gene therapy in a pre-clinical model of CDHR1-associated retinal degeneration - a hitherto untreatable, blinding disorder. Deep phenotyping of the Cdhr1-/- murine model demonstrated severe early deficits in cone and rod photoreceptor function followed by progressive photoreceptor cell death - recapitulating CDHR1 cone-rod dystrophy. Two AAV8 vectors expressing the full-length human CDHR1 coding sequence were designed, manufactured, validated and titrated in vivo. Sub-retinal injection of AAV8.GRK1.CDHR1.pA (1.5 x 108 vg) in Cdhr1-/- mice at 3-4 weeks of age resulted in functional rescue of cone and rod photoreceptors with improved response amplitudes and decreased implicit times to 12-months post-injection; a slowing of photoreceptor cell death with regeneration of full-length photoreceptor outer segments confirmed on ultrastructural analysis to 21-months post-injection; and a behavioural rescue effect on photopic and scotopic optomotor testing, sustained to 21-months post-injection. AAV8.GRK1.CDHR1.pA at 1.5 x 108 vg appeared safe in C57BL/6J mice by the same outcome measures to 22-months post-injection. Genetic prevalence estimates presented herein suggest more than 200,000 affected individuals worldwide. We characterised 146 individuals with biallelic variants in CDHR1, describing the retinal phenotype, natural history, genotype-phenotype associations, with 25 novel pathological sequence variants. This is the first therapy shown to improve retinal structure and function in a pre-clinical model of CDHR1-associated retinal degeneration. A patent to protect the vectors described herein has been filed on behalf of the University of Oxford, with commercial agreements obtained in preparation for an onward phase 1 clinical trial.

Mitochondria are highly dynamic organelles found in most eukaryotic cells, with a fundamental role in the generation of cellular energy through oxidative phosphorylation (OXPHOS). Critical for their function, mitochondria have retained their own genome the mitochondrial DNA, mtDNA. In mammals, replication of mtDNA is ensured by the DNA polymerase POLγ, which is composed by one catalytic subunit POLγA and two accessory subunits POLγB. Mutations in the nuclear-encoded POLG gene, coding for POLγA, are a common cause of human disease leading to a spectrum of disorders characterised by mtDNA instability, thus compromising mitochondrial function. Despite being relatively frequent, the molecular pathogenesis of POLG-related diseases is poorly understood and efficient treatments are missing, partly due to the lack of relevant in vivo models. Here, I describe the generation of two mouse models: 1) the PolgA449T/A449T mouse, which reproduces the A467T change, the most common human recessive mutation of POLG and 2) the PolgWT/Y933C mouse, which reproduces the Y955C change, the most common human dominant mutation of POLG. I focused on the use of the PolgA449T/A449T mouse and complementary in vitro techniques to provide insights into the molecular pathogenic mechanism of this POLG mutation. I describe the data showing that the mouse A449T mutation impairs DNA binding and mtDNA synthesis activities of POLγ, leading to a stalling phenotype. Most importantly, the A449T mutation also strongly impairs interaction with POLγB, the accessory subunit of the POLγ holoenzyme. This allows the free POLγA to become a substrate for LONP1 protease degradation, leading to dramatically reduced levels of POLγA in A449T mouse tissues, with consequences for the pathogenesis of the disease. In the second part of the dissertation, I explore a gene therapy approach for mitochondrial diseases associated with mutations in nuclear-encoded genes. In particular, I test the use of a novel adeno-associated virus (AAV) capsid (PHP.B) as a gene therapy platform to ameliorate the neurological symptoms of a pre-clinical mouse model of mitochondrial disease, the Ndufs4 knockout (Ndufs4-/-) mouse. A single injection with AAV-PHP.B to express the human NDUFS4 in Ndufs4-/- mice, improved lifespan, body weight gain, motor coordination and several molecular and histological features of the brain. These data provide promising proof-of-concept for the use of AAV-mediated gene therapy as a therapeutic option for the number of patients with, currently incurable, mitochondrial disease.

Introduction: Cardiac conduction system (CCS) dysfunction contributes significantly to morbidity and mortality associated with various pathologies, including heart failure. Evidence suggests that remodelling of CCS gene expression, particularly in ion channels regulating its function, are a leading cause of CCS dysfunction. Despite the emergence of these promising therapeutic targets, few methods are available to target the CCS for genetic modulation. Viral vectors offer a powerful method for gene delivery, and AAV9 has been shown to target the working myocardium efficiently. We aimed to utilise the cardiotropic AAV9 vector to modulate gene expression in the CCS from a single systemic intravenous dose, and spatially restrict transgene expression via tissue specific promoters. Doing so would provide a means of validating therapeutic targets of CCS disease, and offer a targeted means of correcting function. Methods: Firstly, tissue specific promoter candidates for use in a CCS specific gene therapy vector were characterised. Two genes where expression is restricted to the CCS in the adult mouse heart were selected; HCN4, a key pacemaking ion channel responsible for If, and KCNE1, which in combination with KVLQT1 regulates IK,s. We also selected the cardiac specific cTnT promoter and the ubiquitous CMV promoter as a positive control. Promoters were cloned upstream of GFP to provide quantitative information on their transcriptional activity. All promoters were first tested in vitro in a sinoatrial node-like ESC derived cell line (Shox2) to assess transcriptional strength, and in off-target cells including HEK293, HIH-3T3 to assess tissue specificity. To enhance transcriptional strength, the 380 bp CMV enhancer was appended upstream of each promoter. After in vitro validation, the most promising promoter constructs were packaged in AAV9 for in vivo testing. Results: All CCS tissue specific promoter candidates showed relatively high activity in Shox2 cells, and low activity in all off target cells. The CMV enhancer increased transcriptional strength modestly in Shox2 cells, with only the CMVe 0.8 kb KCNE1 construct reaching significance. Some evidence of promiscuous activity in HEK293 and NIH-3T3 cells was also noted with CMVe. AAV9-CMV induced high level transgene expression in all tissues examined in vivo, particularly the heart and liver. The SAN was effectively targeted by AAV9, with a transduction efficiency of 65% using AAV9-CMV. CCS specific promoter candidates conferred no detectable transgene expression in any tissues examined. AAV9-cTnT was highly active in the cardiac ventricles and atria, and achieved a transduction efficiency of 7.23% in the SAN. Conclusions: A single systemic intravenous injection of AAV9 is an effective means of targeting the SAN in vivo. Further work is required to elucidate the mechanisms surrounding fine patterning of gene expression in the CCS towards development of potent tissue specific promoter constructs.

A recent analysis of blindness and vision impairment trends over the past 30 years by the Vision Loss Expert Group for the Global Burden of Disease Study revealed that glaucoma remains the leading cause of irreversible blindness worldwide in an ageing population. The mechanisms underlying retinal ganglion cell degeneration in glaucoma are not fully understood. This thesis examines two key features of glaucomatous neurodegeneration, namely impairment of optic nerve axonal transport and dysregulation of aqueous outflow. First we demonstrate the continued need to develop new diagnostic and therapeutic approaches for glaucoma by assessing rates of glaucoma progression in the local Cambridge population. We show there is still a clear rate of progression despite tertiary level access to the best treatments currently available. We describe a new non-invasive tool to assess and quantify the dysregulation of aqueous outflow. Despite being a key therapeutic target in current clinical practice, our understanding of this system is derived largely from post-mortem studies, indirect estimates or, more recently, invasive techniques that alter the physiology of aqueous outflow and are limited in the information they provide. There are currently no non-invasive techniques available that directly visualise detailed aqueous outflow in vivo. We use a non-invasive imaging modality developed to image the anterior microcirculation of the human eye - haemoglobin video imaging - to examine aqueous outflow into the episcleral venous circulation in detail, and to observe and quantify alterations in aqueous outflow following treatments that aim to selectively target this pathway. We then evaluate a combined receptor-ligand gene therapy onstruct which overexpresses brain derived neurotrophic factor and its receptor tropomyosin kinase B on optic nerve axonal transport and retinal ganglion cell function. We use a rat glaucoma model and a humanised tauopathy model of reduced axonal transport and showed combined overexpression is more effective in stimulating axonal transport than either receptor administration or ligand administration alone. We also demonstrate functional recovery in the glaucoma model. Furthermore, we offer promising evidence there may be an improvement in short term memory loss in the tauopathy model, suggesting that this strategy to target intrinsic mechanisms to improve neuronal function and facilitate repair may be relevant to multiple neurodegenerative diseases. We then assess whether manipulating the degenerative process could also be associated with optic nerve regeneration in experimental glaucoma. We demonstrate for the first time the glaucomatous optic nerve also has the potential to regenerate using strategies previously only used after traumatic injury.

The field of gene therapy has changed drastically since the first human gene transfer experiment conducted by Rosenberg et al., over 30 years ago. The field's increasing popularity has brought to light the ever-increasing demand for more efficient and cost-effective solutions to the manufacturing of cell and gene therapies. The need to address these issues has highlighted the numerous challenges that the industry currently faces when striving to develop novel gene therapies. The objective of this research was to investigate improving achievable outputs of cell manufacture for two of the most used adherent-based producer cell lines, HEK293 and HEK293T, with further studies examining the impact on lentiviral vector outputs. Investigations into improving cell culture outputs resulted in the generation of a standardised culture process with predictable, scalable, high yield outputs with a lower cost production. Standardisation of the cell culture process comprised of the implementation of set seeding densities, set culture times, set reagent volumes and defined mixing procedures, all of which led to the creation of a standard operating procedure for adherent HEK culture for use within the Centre for Biological Engineering at Loughborough University. Consistent cell densities of 3.34x105 ± 8.40% cells/cm² for HEK293T and 2.06x105 ± 13.23% cells/cm² for HEK293 were achieved, with the low cost of output cells (HEK293T) ranging between £4.15x10-5 and £4.98x10-5 per individual passage. Furthermore, it was found that the more cells that were produced within a standardised process, the lower the cost to create those cells, with a 42% reduction in cost of creation being observed. Qualities such as these would prove extremely beneficial in the development of scalable clinical grade manufacturing processes. Further studies aimed to identify and address key areas of variation within the transfection processes and examine factors that could influence the point of infection for transfection and the impact on lentiviral vector outputs. Factors such as type of production cell used, the seeding density used, length of incubation prior to transfection, phase of cell cycle and presence of serum at point of transfection were investigated for their impact on transfection success. Additional factors were investigated with the aim of assessing their impact on lentiviral transduction success. The factors being assessed were the type of filter membrane used for clarification of viral supernatant, presence of lentiviral infection enhancing polymers, and volume of viral supernatant added to cells. Baseline investigations generated titres of 7.02x10³ TU/ml in HEK293 cells. Post experimentation, improved titres of 4.67x10⁴ TU/ml and 4.00x10⁵ TU/ml were achieved for HEK293 and HEK293T cells, respectively. Batch to batch variability in lentiviral titres are a known phenomena within the industry, especially with the use of anchorage dependant production systems. This variation was also observed within this body of work and emphasised the current difficulties being faced when developing accessible, scalable, and standardised manufacturing procedures for the generation of viral vectors using producer cell lines.

The amyloid cascade hypothesis suggests that overproduction and accumulation of the neurotoxic amyloid beta (Aβ) peptide plays a central role in Alzheimer's disease (AD). Aβ is generated from the sequential cleavage of the amyloid precursor protein (APP) through the β-secretase pathway. Several isoforms of APP exist: APP695, APP751, and APP770, generated through alternative splicing of exons 7 and 8. Alternative splicing to produce APP695 involves exclusion of exon 7, encoding the Kunitz protease inhibitor (KPI), and exon 8. Studies show up-regulation of KPI-containing APP isoforms (APP770 and APP751) in AD and increase Aβ deposition, whereas APP695 is down-regulated. Furthermore, the AICD (Amyloid Intracellular Domain) which is preferentially produced from APP695 can translocate to the nucleus and act as a transcription factor for genes such as Neprilysin which acts to degrade Aβ. Therefore, APP695 is theorised to be protective against AD, with its loss coinciding with disease progression. This study sought to design Phosphorodiamidate morpholino oligomers (PMOs) specific to exons 7 and 8 of APP, targeting exonic splicing enhancer (ESE) sites, sterically hindering their binding, and modulating alternative splicing, thus increasing APP695 levels. In addition, this study assessed the efficacy of these PMOs and examined the protective effects of APP695 and its cleavage products by analysing downstream effects of splicing modulation. Results demonstrate that PMOs successfully induced alternative splicing events, modulating APP at the RNA and protein level, and significantly increasing APP695. Western blotting data confirmed the presence of the AICD peptide in protein samples from PMO transfected cells. Following this, qPCR data confirmed that PMO transfected cells show increased expression of key Aβ clearing proteins such as Neprilysin. The data collected in this thesis reinforces the theory that APP695 is protective against AD through the effects of the AICD. This strategy emerges as a potential therapeutic for AD treatment.

Adeno-associated virus (AAV) is a viral vector that can be used to deliver therapeutic genes to diseased cells in the eye. Whilst some consider the eye to be an immuneprivileged organ, recent reports have begun to suggest that injection of AAV may elicit local and systemic immune activation. The objective of this thesis was to further our understanding of the nature of the immune response to intraocularly-delivered AAV vectors, and to develop strategies to achieve repeated gene transfer (i.e. successful follow-on injections) to the inner retina. A literature review was conducted into the role that pre-existing and induced neutralising antibody responses may play in the context of ocular gene therapy. This concluded that a multifaceted/comprehensive approach may be required to circumvent anti-capsid humoral immunity and enable repeated gene transfer in humans. As such, two possible components of this strategy were tested in this thesis. First, a capsid mutagenesis strategy was tested in which certain capsid residues, termed phosphodegrons, were mutated to attenuate cytosolic degradation of AAV. This was to enable robust transduction of the murine retina at low vector dose and thereby circumvent the generation of neutralising antibodies that is associated with injection of high doses of AAV. In fact, it was found that phosphodegron mutant AAV serotype 2 (AAV2) resulted in higher levels of immune activation across all tested parameters vs. wild-type AAV2, including neutralising and total AAV binding antibody levels, microglia activation, Muller glia activation, CD4+ and CD8+ T-cell infiltration into the retina, splenic germinal center B-cell activation and class-switching, and activation of conventional dendritic cells. Mechanistically, it was shown that the capsids demonstrated reduced binding affinity to the AAV2 primary receptor, heparan sulphate proteoglycan, and as such it was suggested that reduced sequestration of virions in the heparan-rich inner limiting membrane may increase cellular infection levels in the retina, thereby increasing overall immune activation. The phosphodegron mutant capsids also demonstrated that a slight escape from anti-AAV2 neutralising serum, suggesting they may be more refractory to pre-existing neutralising antibodies in seropositive patients. Second, an assessment was made as to whether perioperative administration of immunosuppressive glucocorticoid steroids would circumvent immune responses to AAV. It was found that prednisolone was effective at reducing anti-AAV neutralising and total AAV binding antibody titres. This effect may have occurred via the inhibition of splenic germinal center B-cell reactions and class-switching of immunoglobulin antibody isotypes, and reducing follicular helper T-cell levels. Despite the reduction in NAb and TAb titres, it was found that perioperative prednisolone administration was insufficient to enable repeated gene transfer following a second eye injection of AAV2. Interestingly, repeated bilateral AAV2 injections may have resulted in a slight reduction in electrophysiological activity in the retina compared to mice who received two control PBS injections. In conclusion, this thesis highlighted novel aspects of phosphodegron mutant AAV2 immunobiology and identified a possible utility for the administration of prednisolone as a tool for enabling repeated gene transfer of AAV gene therapy vectors.

Gene therapies are being developed for rare genetic diseases and offer hope for conditions with few treatment options. Surfactant protein B (SP-B) deficiency is a rare genetic disease of the lung causing severe respiratory distress in neonates and is fatal within the first months of life. Recombinant adeno-associated virus (rAAV) is being investigated for delivery of functional SP-B to the lungs of SP-B deficient neonates, to produce surfactant for secretion onto the lung surface to increase chances of survival. To screen AAV serotypes for efficient targeting of alveolar type II (ATII) cells of the lung parenchyma, rAAV5, rAAV6.2 and rAAV9 expressing reporter transgenes were delivered to murine lungs via intranasal instillation. Serotype 5 mediated the highest transduction levels in the target ATII cells (19.4 %, p< 0.05 versus naive). The in situ hybridisation method RNAscope was adapted to include a DNase digestion step to allow for detection of transgene mRNA expression after rAAV delivery and showed that the hCEFI promoter mediated the highest expression levels in ATII cells (11.1 transgene mRNA copies per cell, p< 0.05 versus rAAV5.SP-B MG and p< 0.0001 versus all other groups). Therefore, rAAV5.hCEFI.SP-B was the rAAV vector selected for use in the murine lung. To validate the vector for human lung delivery, eight AAV capsids were screened in human models of the lung parenchyma: precision cut lung slices (PCLS) generated from lung resections and lung bud organoids (LBO) generated from human embryonic stem cells. The rAAV5 vector did not transduce the PCLS model, or the LBO after microinjection to the lumen; but vectors rAAV6, rAAV6.2 and rAAV6.2FF showed distinct levels of transduction in both models (p< 0.005 versus naive), indicating their suitability for gene delivery to the human lung parenchyma and highlighting the limitations of the murine model for AAV serotype selection in lung. Reporter gene expression using the CMV promoter was more robust than the hCEFI promoter in both the PCLS (hCEFI expression not detected) and LBO model (about 5-fold), but hCEFI mediated distinct levels of expression long-term. To test the capability of rAAV6.2 to mediate therapeutic protein expression in a human model of the SP-B deficient lung, embryonic stem cells were gene edited to incorporate the most common disease-causing 121ins2 mutation in SP-B deficiency. Subsequently, LBOs were generated from this edited cell line by sequential differentiation and delivery of rAAV6.2.CMV.SP-B resulted in expression of mature SP-B protein, highlighting the translational potential of the selected rAAV vector.

Dominant mutations in RHO (rhodopsin) are the most common cause of autosomal dominant retinitis pigmentosa (RHO-adRP). RHO-adRP causes progressive loss of rod cells, followed by cone cells, leading to blindness. This disease is a candidate for CRISPR gene editing, as reduction of mutant rhodopsin is associated with phenotypic rescue. This thesis optimises all-in-one CRISPR/Cas9 and CRISPRi adeno-associated viruses (AAVs) and explores their potential as RHO-adRP gene therapy vectors. Following construct optimisation in vitro, a CRISPR/Cas9 and CRISPRi plasmid carrying SaCas9 and dSaCas9.KRAB, respectively, were generated. Both demonstrated knock down of endogenously expressed EGFP in a transgenic cell line. For allele-specific knock down of RHO, the CRISPR constructs must target a region unique to the mutant allele. Bioinformatic screening of RHO identified seven non-pathogenic single nucleotide polymorphisms (SNPs) that were targetable with SaCas9. Two of these SNPs could be targeted allele-specifically, with high CRISPR/Cas9 gene disruption and CRISPRi gene repression rates of 38.5-73.2 % and 43.1-65.8 %, respectively. In heterozygous patients, targeting the SNP on the mutant RHO offers a mutation-independent allele-specific targeting strategy. As a proof-of-concept for targeting rod cells, the AAV were subretinally injected into Nrl-EGFP mice, in whom enhanced green fluorescent protein (EGFP) expression is limited to rod photoreceptors. To restrict Cas9 expression to rods cells only, a novel 154 bp rod cell-specific promoter was identified. This shortened promoter from the PDE6B gene was active in human retinal explants, the human retinoblastoma-derived cell line Y79, and drove strong rod cell-specific expression in mice. The CRISPRi AAV was unable to repress EGFP, despite high transgene expression. The CRISPR/Cas9 AAV however, drove strong EGFP disruption. It produced 4.1 % EGFP indels, resulting in a reduction of EGFP mRNA and fluorescence of 55.3% and 36.5 %, respectively. The all-in-one CRISPR/Cas9 AAV is therefore able to drive strong gene disruption in rod cells in vivo, making it a useful vector for the treatment of RHO-adRP. This thesis describes in detail the development of these all-in-one CRISPR/Cas9 and CRISPRi AAV vectors from proof-of-concept to in vivo study and advances the potential for using such therapies in future treatments of RHO-adRP.

Adeno-associated virus (AAV) shows great promise for gene therapy, however scalability, yield and quality remain significant issues. The most common 'helper-free' method used to produce recombinant AAV (rAAV) for pre-clinical/clinical application is based on plasmid transfection of HEK293 cells to provide all the components required for rAAV replication. While versatile and easy to adopt, the helper-free method is limited by a lack of scalability and vector yield. Here we describe novel rAAV manufacturing strategies using 'helper' adenoviruses that self-inhibits its major late promoter (MLP) to truncate its own replication. Modification of the adenovirus MLP in situ to provide inducible repression has not previously been demonstrated, primarily because the virus DNA polymerase coding sequence is in the opposing DNA strand. We show that strategic insertion of tetracycline repressor binding sites into the MLP and encoding the tetracycline repressor under its transcriptional control allows for normal adenovirus replication in the presence of doxycycline, but only enhanced genome amplification and early gene expression (important for the 'helper' functions) in its absence. Using this negative feedback system, we developed a complete transfection-free 'self-repressing' helper-vector approach to produces AAV particles. We demonstrate delivery of adenoviral helper functions, the rAAV transfer genome, and the AAV genes including rep, known to be toxic to adenovirus, to yield up to a 1000-fold increase in infectious rAAV2 compared to the helper-free plasmid system. We further extended this approach to produce rAAV5, -6, -8, -9 vector serotypes, and in all cases, yielding significant increases in intact particles over the standard manufacturing approach with over 8 log reduction in contaminating adenoviruses. To reduce manufacturing complexity and encapsidation of illegitimate DNA within rAAV particles, we also explored a unique strategy to manufacture rAAV vectors by propagating rAAV particles using self-silencing adenoviral vectors expressing the AAV components. This included a trans-pseudotyping or 'capsid-exchange' method to generate rAAV-5, -6, -8, and -9 from rAAV2 particles. We propose these as high-yielding contaminant-free production systems suitable for scalable rAAV manufacture. Lastly, in exploring the use of the self-silencing adenovirus to promote rAAV production from mammalian cells stably encoding the AAV genes, we provide direct evidence that factor(s) expressed from the MLP during adenovirus replication was required to support CARE-dependent amplification, a unique mechanism that results in 100-1000 fold DNA amplification of the integrated AAV genes, and a prerequisite for high-titre rAAV production from the stable AAV packaging cells.

Recombinant lentiviral vectors (LV) are an attractive vehicle for gene delivery to the lung due to their ability to achieve stable, long-term, expression and the opportunity to add glycoprotein pseudotypes for efficient lung cell-targeting. Two LV pseudotypes derived from sialic acid- binding Sendai virus and influenza virus, referred to as F/HN and Rostock respectively, can facilitate efficient in vivo transduction of the murine airway and were compared alongside a new influenza-based pseudotype (Shanghai) which, in contrast to F/HN and Rostock, is based on a strain that infects humans. Lectin staining confirmed a sialylated receptor distribution that is different between murine and human lung, with a greater abundance of a2,6 and relative lack of an a2,3 subtype in the human airway, such that transduction of mouse lung is likely a poor predictor of performance in humans. Ex vivo transduction of murine precision cut lung slices (PCLS) recapitulated results from in vivo experiments. However, PCLS from resected human lung did not contain airways, and therefore cultures of human airway epithelial cells, maintained at an Air-Liquid Interface (ALI), were used to model transduction of the polarised airway epithelium. Different cellular source materials and ALI culture protocols showed differences in the abundance of human sialylated receptor subtypes available, as well as subsequent transduction efficiency by F/HN pseudotyped LV. Importantly, transduction of human ALI cultures was more efficient using vectors based on human as opposed to simian immunodeficiency virus. A greater transduction efficiency was observed for Shanghai pseudotyped LV compared with Rostock but, overall the F/HN pseudotype demonstrated superior efficiency, transducing all human airway cell types characterised. This was consistent with the greater affinity of the F/HN pseudotype for the a2,3 sialylated (N-acetyllactosamine) receptor present in all models investigated, and also explained the broad range of pulmonary cell-types (including the repopulating/progenitor cells of murine airway) transduced in all models investigated. These data support the strong translational potential of the F/HN pseudotype.

Dose-dependent inflammatory responses in adeno-associated virus (AAV)-mediated retinal gene therapy trials have restricted the use of high vector doses and impeded the ability to enhance therapeutic efficacy. Characterising the nature of this response, and devising methods to overcome it, are thus becoming increasingly important in the clinical application of gene therapies. In the experiments herein, I investigated the composition, timing, and magnitude of the cellular immune response to retinal gene therapy in mice using quantitative flow cytometric analysis. I observed a significant increase in retinal leukocytes from 14 days after subretinal injection of an AAV8 vector expressing green fluorescent protein (GFP). These consisted of a constant population of resident microglia and infiltrating peripheral leukocytes including macrophages, natural killer cells, CD4 and CD8 T cells, and natural killer T cells. An identical vector at an equivalent dose encoding a human therapeutic transgene elicited a diminished response compared to paired injections with a GFP vector. However, the proportions of the leukocyte populations remained similar, suggesting a common mechanism of immunity. In addition to the induction of a cell-mediated immune response, innate detection of AAV vectors may directly restrict transgene expression. Preliminary evidence suggested in vitro AAV transduction may be limited by the anti-viral nucleic acid sensors Toll-like receptor 9 and cyclic GMP AMP synthase, suggesting possible detection of transgenic AAV genomes. I also showed that the immunomodulatory drug hydroxychloroquine can significantly enhance AAV transduction in vitro, ex vivo, and in vivo, with indirect indication of the partial antagonism of TLR9. The lack of structural change in the retina following subretinal hydroxychloroquine delivery indicates that it could be a safe adjunct for improving retinal gene therapy efficacy.

Tumour cells exhibiting the Warburg effect rely mostly upon aerobic glycolysis for ATP production and have a notable addiction to anaplerotic use of glutamine for macromolecular synthesis. This strategy maximises cellular biosynthetic potential and provides an attractive anabolic environment for viral infection. Here, we evaluate infection of highly permissive cells (A549 cells) and poorly permissive cells (SKOV3) with wildtype adenoviruses, and the oncolytic chimeric adenovirus enadenotucirev (EnAd) to probe the role of metabolism on the success of oncolytic virotherapy with EnAd. All adenoviruses caused an increase in glucose and glutamine uptake along with increased lactic acid secretion in their host cells, mirroring the canonical Warburg effect phenotype. Through identifying the IC₅₀ of a panel of cancer cells, we identify a correlation showing that cells which rely upon aerobic glycolysis tend to be more permissive to EnAd oncolysis. The virus lifecycle of both wildtype adenoviruses and EnAd were critically dependent on exogenous glutamine. Virus activity in glutamine-free cells was rescued with exogenous membrane-permeable α-ketoglutarate, but not pyruvate or oxaloacetate, suggesting an essential role for reductive carboxylation in glutamine usage. We propose that this unusual glutamine metabolism is beneficial for the production of biosynthetic intermediates, particularly under conditions of hypoxia or decreased cellular respiration as is often seen during adenovirus infection. Counterintuitively, restricting glycolysis using 2-deoxyglucose or by limiting glucose supply strongly improved virus activity in both cell types. Antagonism of glycolysis also boosted EnAd replication and transgene expression within human tumour biopsies, and in xenografted tumours in vivo. We further bolstered this finding through weaning SKOV3 cells off of glucose. This process created a cell line which is highly permissive to EnAd infection and lysis. Finally, we engineered an "armed" EnAd expressing the glycolysis inhibitor TIGAR which was able to lyse cells whilst expressing TIGAR without causing acidification of the media. These findings emphasise the importance of metabolism on productive viral infections which is of particular significance for viruses restricted to replicate in tumours, that already display aberrant metabolic phenotypes.

Background: Congenital myasthenic syndromes (CMS) are a group of inherited disorders that affect signal transmission at the neuromuscular junction (NMJ) and share the clinical feature of fatigable muscle weakness. Downstream of tyrosine kinase 7 (DOK7) is a cytoplasmic-adaptor protein that amplifies the signalling from the muscle-specific receptor tyrosine kinase (MuSK), which is responsible for the formation and stabilisation of the NMJ. It has recently been shown that overexpression of DOK7 protein in muscles, generates enlarged NMJs in mice, which are very efficient in signal transmission, have no reported detrimental effects and can ultimately lead to a complete recovery for disease model mice. Aim: To study novel therapies for CMS based on DOK7 overexpression. Results: By using an adeno-associated virus system (DOK-AAV) we generated enlarged NMJs in wild-type mice to further investigate the morphology and electrophysiology of these synapses. The DOK7-AAV gene therapy was shown to generate enlarged synapses that improved the phenotype of two animal models of CMS. In addition, 5431 small molecules from a library of muscle-specific compounds were used in screening experiments to find potential small molecules that could upregulate DOK7. Confirmatory results were performed on an assay of biological activity and revealed two potential small molecule hits. They will provide the basis for a more detailed analysis of the effects of DOK7 upregulation in cell culture biological assays and ultimately for in vivo testing. Conclusion: These studies demonstrate that DOK7 upregulation would benefit differing CMS subtypes and could potentially improve signal transmission in other NMJ-associated disorders.

Transient dedifferentiation of cardiomyocytes has been established as a key mechanism behind the enhanced proliferation and cardiac regenerative capacity observed in lower vertebrates. In contrast, mature mammalian cardiomyocytes exist in a terminally differentiated state such that regeneration is significantly impaired. It has been previously demonstrated that somatic cells can be reprogrammed temporarily to a proliferative, dedifferentiated state in vitro and in vivo through transient overexpression of Oct3/4, Sox2, Klf4 and c-Myc (OSKM). However, the response of cardiomyocytes to transient OSKM expression has not previously been explored. In this thesis, the effects of adenoviral vector mediated OSKM delivery to cardiac cells in vitro and in vivo were investigated. Forced expression of OSKM in cultures of neonatal rat and mouse cardiomyocytes induced rapid dedifferentiation as evidenced by sarcomere dis-assembly and downregulation of cardiomyocyte-specific genes (Myh6, Myh7). This induced dedifferentiation was associated with an increase in cell cycle activity consistent with an enhanced proliferative capacity. The initiation of reprogramming was further confirmed by the expression of early stage reprogramming markers (ECad) however, late-stage reprogramming markers remained silenced and the generation of pluripotent stem cells was not observed. Furthermore, parallel to decreases in exogenous OSKM expression, partially reprogrammed cells regained a contractile cardiomyocyte-like morphology and a gene expression profile consistent with their redifferentiation towards functional cardiomyocytes. The effect of ectopic OSKM expression was also investigated in vivo through direct intramyocardial injection of the same adenoviral vector. Within 6 days after injection, upregulation of endogenous pluripotency genes (endogenous Oct3/4 and Gdf3) was observed indicating the activation of a reprogramming-like response within the adult mouse myocardium. The increased expression of such genes was maintained only temporarily consistent with a transient reprogramming response which was further evidenced by the absence of teratomas. Finally, the capacity of an OSKM expressing adenoviral vector to induce this transient reprogramming response in a mouse model of myocardial infarction was confirmed. In conclusion, this work reports the first evidence of adenoviral mediated transient reprogramming of cardiac cells both in vitro and in the healthy and injured myocardium in vivo. Furthermore, this study was the first to explore adenoviral vectors for in vivo OSKM delivery and highlighted challenges associated with this approach. Further work is necessary to understand the source and fate of partially reprogrammed cells as well as determine whether this strategy can provide any therapeutic benefit in the heart. However, it is hoped that the results presented here can help inform future research into the development and applications of transient reprogramming-based strategies.

Neural networks are at the core of the brain’s ability to compute complex responses to our external environment. Clinically, network dysfunction is emerging as a key component of several psychiatric and neurodegenerative disorders such as Alzheimer’s disease or schizophrenia. However, our ability to precisely and safely manipulate neural networks for research and deliver network-specific therapy remains limited. To address this problem, our lab recently developed a monosynaptically restricted Self-Inactivating Rabies virus (SiR) which enables the targeting of neural circuits without cytotoxicity. To expand the scope of SiR we further developed the technology in two directions: A) By incorporating the CRISPR/CAS9 gene-editing machinery into the SiR genome to successfully edit endogenous loci in vitro and in vivo. B) By designing an improved second generation SiR virus (SiR 2.0) which applies the same SiR technology to a challenge rabies strain (CVS-N2C). SiR 2.0 demonstrates increased neurotropism, increased trans-synaptic transfer efficiency and markedly decreased immunogenicity compared to the SiR 1.0 vector. These advancements expand the scope of SiR viruses to be used in the genome-editing of circuits in vivo. A combined SiR 2.0 CAS9 virus, in physiology, allows us to investigate the roles of genes within circuits in the brain function of live animals. For therapy, it paves the way for the rabies virus’ potential use to edit disease-related genes in dysfunctional circuits. Despite the circuit-basis of many neurological disorders, existing gene therapy vectors are not circuit specific. In addition, the practical difficulties of delivering therapeutic agents at high doses into the central nervous system exacerbates our inability to achieve high therapeutic loads into affected circuits. In contrast, a SiR 2.0 CAS9 virus would, following injection into peripheral organs, trans-synaptically spread into desired circuits of the central nervous system that are affected in neurological disease (e.g. networks demonstrating pathological protein propagation in neurodegenerative disorders) and edit disease-related genes. Lastly, our interest in network-level pathological protein propagation also led us to investigate the biology behind this observation. Due to additional evidence that a significant number of other proteins in physiology also show interneuronal movement, we hypothesised that perhaps this is an overlooked phenomena in neurobiology which could have key implications.

Endothelial cells (ECs) have a well-established role in the maintenance of vascular physiology. Moreover, ECs participate in many physiological processes such as angiogenesis, blood homeostasis, inflammatory response, lipid metabolism and many more. Angiogenesis is one of the most studied functions of endothelium, where ECs participate in the formation of the new blood vessels. Additionally, ECs can also exhibit a form of plasticity called endothelial-to-mesenchymal transition (EndMT), which is characterized by the loss of endothelial-specific morphology and markers and acquisition of mesenchymal-like phenotype. Both angiogenesis and EndMT can be regulated by physiological cues, such as inflammation and haemodynamic forces, as well as number of molecular stimuli, such as TGF-β. Interestingly, recent evidence suggests that angiogenesis and EndMT can be orchestrated by miRNAs. This thesis aims to address the hypothesis that miR-148b is a modulator of angiogenesis and endothelial cell plasticity in the form of EndMT. The preliminary high-throughput miRNA screen has identified miR-148b as a strong enhancer of HUVEC proliferation. Subsequent bioinformatic analysis and validation demonstrated that TGFB2 and SMAD2 are direct targets of miR-148b in ECs. Further experiments using gain- and loss-of-function approaches demonstrated that miR-148b regulates EC function. Specifically, overexpression of miR-148b enhanced EC migration, proliferation and in vitro angiogenesis, whereas its inhibition promoted EndMT, decreasing the expression of CD31 and VE-Cadherin and elevating collagen 1. Furthermore, inflammatory cytokine challenge decreased miR-148b levels in ECs, promoting EndMT, via upregulation of SMAD2, and enhancing reactive oxygen species production, all of which were abrogated by exogenous miR- 148b. Finally, in a mouse model of skin wound healing, delivery of miR-148b mimics promoted wound angiogenesis and accelerated wound closure. In contrast, inhibition of miR-148b enhanced EndMT in wounds impairing wound closure, which is reverted by SMAD2 silencing. Together, the data in this thesis supports the hypothesis that miR-148b regulates angiogenesis and endothelial cell plasticity, and provides the evidence that miR-148b upregulation enhances both in vitro and in vivo angiogenesis, while its knockdown promotes EndMT. This thesis demonstrates for the first time that miR-148b could be a key factor controlling EndMT and vascularization, thus opening a new avenue for therapeutic application of miR-148b in diseases that require vascular and tissue repair.

Mutations in the rhodopsin gene are one of the most common causes of autosomal dominant retinitis pigmentosa (ADRP) and there is at present no treatment available. In this thesis, three potential treatments were developed and tested in a RhoP23H/+ knock-in mouse model. In the first, animals were housed in red cages which acted as long-pass light filters. This resulted in a significant slowing in the rate of both rod and cone photoreceptor degeneration as determined anatomically using non-invasive in vivo imaging and immunohistochemistry, and functionally by dark- and light-adapted electroretinography. In the second, efficient adeno-associated viral (AAV) vectors expressing wild type human rhodopsin were manufactured and injected subretinally in RhoP23H/+ mice to test the hypothesis that overexpression of the normal protein might overcome the effect of the P23H mutant. No structural or functional rescue effect was however observed. The final treatment strategy involved development of artificial mirtrons to suppress expression of endogenous rhodopsin. Mirtrons are novel RNA interference effectors that are spliced from mRNA transcripts. Both efficient mirtrons and a codon modified mirtron-resistant version of the human rhodopsin coding sequence were designed and validated in vitro. These elements were then combined under the control of a single rhodopsin promoter within AAVs as knockdown/replacement gene therapy vectors for ADRP. Subretinal injection in the RhoP23H/+ mouse resulted in mouse-to-human rhodopsin mRNA replacement, and this was associated with a modest rescue effect when injected at a dose of 2x108 gc. This represents the first use of an artificial mirtron to induce gene knockdown in vivo.

The development of new gene therapy products offers ground-breaking opportunities for the treatment of human disease. To provide lasting benefit to patients, the expression of a correct copy of a non-functional gene needs to be life-long, and the maintenance of therapeutic levels of gene expression is a key challenge. Gene therapy for chronic liver disease using recombinant adeno-associated virus, is very encouraging, but as the vector exists as an extra-chromosomal episome, it can be lost during subsequent cell division, especially in paediatric liver disease. In this study, maintenance of transgene expression was investigated through repeated administration of rAAV after inducing immunogenic tolerance to rAAV capsid proteins. Using a liver-specific rAAV2/8, induction of tolerance to the non-self transgene luciferase was successfully demonstrated in mice. However, rAAV2/8 delivery of the AAV8 capsid was unable to induce humoral tolerance to the AAV8 capsid itself. Similarly, a lentiviral directed approach in the liver did not induce tolerance to AAV8 capsid. An alternate method to achieve life-long transgene expression utilised a gene editing strategy called homology-independent targeted integration (HITI), that facilitates insertion of a transgene sequence in a precise location in the genome, allowing, for example, therapeutic transgene expression under the strict control the endogenous promoter of an otherwise defective target gene. A fluorescent reporter cell line was constructed to test the feasibility of HITI and was used to confirm multiple permutations of the integration location and the sequence inserted, each of which offers a distinct therapeutic opportunity. Optimisation of the HITI donor sequence was required to prevent potential false positive expression in the analysis of in vivo HITI applications. One permutation was successful in addressing alpha-1 antitrypsin deficiency in vitro and subsequently in a disease mouse model. Two new quantitative analytical methods, Capture-Seq and CasEnrich, were used to determine the propensity for off-target integration in the models tested. While these next-generation sequencing approaches were optimised for analysis of HITI on- and off-target integration events, they can be readily applied to interrogate the genomic consequences other gene editing and gene therapy approaches. In conclusion, in the absence of a successful protocol for repeat rAAV administration in vivo, the HITI approach provides a promising platform for targeted genomic integration leading to life-long expression of therapeutic transgenes from an endogenous promoter.

Duchenne muscular dystrophy (DMD) is a hereditary X-linked neuromuscular disease resulting from mutations across the DMD gene. The subsequent absence of the encoded dystrophin protein prevents the correct formation of the dystrophin-associated protein complex (DAPC), a structural link between the sarcolemmal actin cytoskeleton and the myofibre extracellular matrix. Dystrophin deficiency compromises myofibre stability giving rise to progressive muscle wasting, the prominent pathological feature of this disease. Over time, muscle deterioration results in loss of ambulation and upper body movement, decline in respiratory and cardiac function and ultimately premature death in the third decade of life. Current treatment is limited to corticosteroid use, which delays clinical manifestations but does not address the underlying aetiology of the disease. In addition, many therapies currently under development address specific mutation types or are reliant upon repeated administration; this serves to reduce their applicability to patients. Genome editing refers to modification of genetic material at a precise genomic locus using special enzymes called customisable endonucleases. Recently, interest in genome editing approaches has increased with the advent of the clustered regularly interspaced palindromic repeats - Cas9 system (CRISPR/Cas9), in which an RNA guide can be used to direct the Cas9 nuclease to cleave the genome at a specific sequence. In the present study a genome editing strategy was designed and evaluated to restore functional dystrophin expression from the endogenous DMD gene. The approach involved modifying intron 1 of the locus to insert dystrophin cDNAs downstream of the endogenous full-length dystrophin promoter and exon 1 elements, and was developed and verified in HEK293T cell and human DMD patient myoblast cultures. The approach was developed to permit full-length, mini-dystrophin and microdystrophin cDNA insertions to be examined. Specific CRISPR/Cas9 and guide RNAs (gRNAs) were designed and assessed for their ability to introduce double strand breaks (DSBs) at the DMD intron 1 locus. An exogenous DNA repair template construct, targeted to the DSB region in intron 1 was designed and further modified to contain a microdystrophin coding cDNA from DMD exon 2 onwards Co-delivery of CRISPR/Cas9, gRNA and exogenous repair template constituents in HEK293T cells, resulted in specific integration of the microdystrophin cDNA at the DMD intron 1 site. This was confirmed at genomic, transcriptomic and protein levels in polyclonal and zeocin-selected clonal cell populations. Application of the genome editing system in human DMD myoblasts also gave rise to evidence of specific microdystrophin cDNA integration, although further refinement of the methodology is required to ensure myoblast survival following positive selection with the zeocin antibiotic. The research reported demonstrates the potential of genome editing the 5' end of the DMD gene and offers a strategy towards permanent corrective gene therapy applicable to almost all DMD patient genotypes.

Introduction: The corneal dystrophies are a group of inherited disorders that affect the shape or transparency of the cornea. The unique qualities of the cornea coupled with their mendelian traits make the them an ideal target for gene therapy. The main aim of this thesis is to investigate the potential to use CRISPR/Cas9 to treat the autosomal dominant TGFBI corneal dystrophies. Methods: Paper II: Plasmid expressing CRISPR/Cas9 targeted to Luc2 was delivered to the cornea via an intrastromal pressure injection. Knockdown of luciferase was measured using an IVIS in vivo imager. The allele-specificity of CRISPR/Cas9 was assessed in vitro using a dualluciferase assay and an in vitro digestion. Paper III: A morpholino targeted to tgfbi was electroporated into the regenerating adult zebrafish tail fin, and the effect of Tgfbi knockdown was measured by the quantification of regenerated tissue. 10X phased sequencing and EBV-transformation of isolated lymphocytes was performed on a patient harbouring a R124H granular corneal dystrophy type II (GCD2) mutation. The phasing information was used to design allele-specific sgRNAs which were then tested using an in vitro digest and targeted resequencing across the target locus in TGFBI. qPCR was used to infer the efficiency of a dual-cut using a combination of sgRNAs. Paper IV: The transduction efficiency of 3 AAV-GFP serotypes in immortalised human corneal cells was assessed using flow cytometry. In vivo transduction of AAV-GFP following intracameral injection was assessed using the IVIS in vivo imager and fluorescent microscopy. AAV-Cas9 and AAV-sgRNA-GFP were co-injected, DNA was extracted from the whole cornea and TIDE analysis was performed to determine efficiency of indels. Results: Paper II: The failure of a mutation-dependent approach to target 20% of TGFBI missense mutations was demonstrated. Comparison of two widely used allele-specific strategies revealed a PAM-specific approach conferred superior specificity than that of a guide-specific approach. The inability of S.pyogenes Cas9 to distinguish between single base pair changes in the guide sequence was confirmed. Paper III: Knockdown of Tgfbi in the regenerating zebrafish tail fin was shown to impair wound healing. The ability to selectively target the mutant allele by means of non-diseasecausing SNPs, which are associated with a PAM on the same allele as the disease-causing mutation, was demonstrated. The addition of a 50:50 ssODN with the ribonucleoprotein complex was shown to significantly increases the frequency of a dual-cut event. Paper IV: AAV-2/9 was shown to transduce all corneal layers in vivo following a single intracameral injection. A dual-AAV-2/9 CRISPR/Cas9 system was shown to generate 25.7% within the whole cornea. Conclusion Cas9 lacks the specificity to discriminate between single base pair mismatches within the 20bp guide sequence, however mutations within the 2bp PAM are much less tolerated. As such, using currently available nucleases a PAM-specific approach is necessary to discriminate between alleles. AAV was demonstrated as a robust vehicle to deliver geneediting reagents to the corneal layers.

Androgens are essential for the maintenance of male health and wellbeing. A disturbance in androgen signalling has been associated with a number of clinically relevant disorders such as cardiovascular disease, diabetes and metabolic disorders as well as infertility. Primarily produced in the testis in males, the actions of androgens are mediated through binding to androgen receptor (AR), a member of the nuclear receptor superfamily of ligand-activated transcription factors. The somatic cells of the testis are known to have a number of key roles in both testis function and development and the Sertoli, Leydig and Peritubular Myoid cells are known to express AR in adulthood. It is through AR that some testicular functions are mediated; for example, the Sertoli cells support of complete spermatogenesis with Sertoli cell androgen receptor knockout (SCARKO) testis demonstrating a halt of spermatogenesis before meiosis. However, how androgen signalling is impacting testicular function through each of the somatic cell types is not yet fully understood. Currently, treatments for male reproductive disorders such as hypogonadism (low androgens) and infertility are limited to treatment of the symptoms; using androgen replacement therapy and in vitro fertilisation techniques. This has been, up until recently, a result of a lack of understanding of the causes of these conditions and a lack of resources able to treat them, with research suggesting that a genetic component may be responsible in a number of cases. However, due to the limited genetic investigation diagnosis of men with male reproductive disorders, the wider understanding of the genetics underpinning male hypogonadism and infertility is incomplete. Developments in technology for the investigation and editing of the genetic code are triggering a surge in the exploration of genetic disorders and, in parallel, into the fields of gene delivery vectors and editing technologies. These technologies will allow an expansion into the knowledge and understanding of genetic disorders whilst simultaneously affording the opportunity to exploit this understanding for the development of therapeutics. There have been a small handful of previous studies using technologies such as viral vectors to target the testicular somatic cells and deliver exogenous transgenes with the purpose of both gene editing and repair, all with varying degrees of success. Here, techniques to introduce and target the Leydig and Sertoli cells were investigated to determine the most appropriate methodology for gene delivery to and manipulation of the testis. Refinement of injections into the interstitial compartment were carried out before introducing lentiviral vectors and targeting of Leydig cells was validated and optimised. Lentiviral vectors are able to permanently integrate into the host cell. Surprisingly, analysis of testis post lentiviral injection determined that the lentiviral targeted Leydig cells began to undergo apoptosis one week post injection and were subsequently cleared from the testis after ten days. Contrastingly, this was not the case when adenoviral vectors were introduced into the interstitial compartment, with Leydig cells continuing to express the delivered reporter transgene and, importantly, not expressing markers of apoptosis, ten days post injection. This would suggest that using adenoviral vectors to target the Leydig cell population in the adult testis would be more appropriate than using lentiviral vectors. Previous studies have successfully used lentiviral vectors to target the Sertoli cells in the adult testis via the introduction of the particles through the efferent duct. However, this can result in damage to efferent duct, resulting in blockages and subsequently the seminiferous tubules. To circumvent this, introduction of the lentiviral particles through the rete compartment of the testis at a range of lower injection pressures was examined and injecting at a lower pressure through the rete testis was found to reduce the likelihood of introducing negative impacts on testicular histology when targeting the seminiferous tubules. Using these refined methods of introducing lentiviral vectors, targeted Sertoli cells stably expressed the delivered transgene for up to one year post injection. Using viral vector delivered transgenes for both the investigation of testicular genetic disorders and for the development of therapeutics has great potential. To explore this potential, we first generated a mouse model in which AR was ablated from both the Leydig and Sertoli cells using Cre/LoxP technology, termed the SC-LC-ARKO. Alongside providing a potential model to 'repair' with viral vectors, the SC-LC-ARKO model also provided an additional model for comparison with other models exhibiting ablation of AR from both single somatic cell types and double somatic cell types. This further enabled a characterisation of the roles of AR in adult testicular function, with results suggesting that loss of AR from more than one cell type results in an additive phenotype when compared to single cell knock outs. Despite providing further insight into the roles of AR in the testis, further analysis of the Cre line used to generate the SC-LC-ARKO model indicated that a small number of Leydig cells were expressing the Cre recombinase, resulting in only a small population of Leydig cells with ablated AR. Considering this, to explore the potential of rescuing Sertoli cell AR using lentiviral vectors, we then utilised an already well characterised Sertoli Cell AR knockout (SCARKO) model. Lentiviral vectors expressing mouse AR and monomeric GFP (moeGFP) downstream of a CMV promoter were generated and injected into the rete testis of WT and SCARKO adult (day 100) males at low pressure. The contralateral testis was injected with a lentiviral vector expressing moeGFP alone (also downstream of a CMV promoter) using the same technique. Analysis of testis sections revealed a reintroduction of AR to Sertoli cells in 100% of SCARKO testis injected with lentivirus expressing mouse AR. As a result of this re-expression of AR in Sertoli cells, 66% of the testis injected with lentivirus expressing mouse AR had evidence of morphologically mature elongated spermatids, indicative of ongoing spermatogenesis. These results suggest that a rescue of the infertility phenotype reported in previous studies of SCARKO testis. Also demonstrated is the reversal of the SCARKO testicular phenotype in tubules targeted by the mAR expressing lentiviral vector. This suggests that absence Sertoli cell AR throughout development does not have a permanent impact on the Sertoli cells capacity to support spermatogenesis in adulthood following rescue of SC AR expression in adulthood. In summary, the results of these studies have provided a refinement in the methodologies for targeting the Sertoli and Leydig cells of the adult testis with viral vectors as well as demonstrating successful rescue of a previously reported mouse model exhibiting infertility through reintroduction of a functional gene. Alongside this, comparisons of AR knockout models have afforded insight into maintenance of testis function through AR.

Rare diseases, when considered as a whole, affect up to 7% of the population, which would represent 3.5 million individuals in the United Kingdom alone. However, while 'personalised medicine' is now yielding remarkable results using recent sequencing technologies in terms of diagnosing genetic conditions, we have made much less headway in translating this patient information into therapies and effective treatments. Even with recent calls for greater research into personalised treatments for those affected by a rare disease, progress in this area is still severely lacking, in part due to the astronomical cost and time involved in bringing treatments to the clinic. Gene correction using the recently-described genome editing technology CRISPR/Cas9, which allows precise editing of DNA, offers an exciting new avenue of treatment, if not cure, for rare diseases; up to 80% of which have a genetic component. This system allows the researcher to target any locus in the genome for cleavage with a short guide-RNA, as long as it precedes a highly ubiquitous NGG sequence motif. If a repair sequence is then also provided, such as a wild-type copy of the mutated gene, it can be incorporated by homology-directed repair (HDR), leading to gene correction. As both guide-RNA and repair template are easily generated, whilst the machinery for editing and delivery remain the same, this system could usher in the era of 'personalised medicine' and offer hope to those with rare genetic diseases. However, currently it is difficult to test the efficacy of CRISPR/Cas9 for gene correction, especially in vivo. Therefore, in my PhD I have developed a novel fluorescent reporter system which provides a rapid, visual read-out of both non-homologous end joining (NHEJ) and homology-directed repair (HDR) driven by CRISPR/Cas9. This system is built upon a cassette which is stably and heterozygously integrated into a ubiquitously expressed locus in the mouse genome. This cassette contains a strong hybrid promoter driving expression of membrane-tagged tdTomato, followed by a strong stop sequence, and then membrane-tagged EGFP. Unedited, this system drives strong expression of membrane-tdTomato in all cell types in the embryo and adult mouse. However, following the addition of CRISPR/Cas9 components, and upon cleavage, the tdTomato is rapidly excised, resulting via NHEJ either in cells without fluorescence (due to imperfect deletions) or with membrane-EGFP. If a repair template containing nuclear tagged-EGFP is also supplied, the editing machinery may then use the precise HDR pathway, which results in a rapid transition from membrane-tdTomato to nuclear- EGFP. Thereby this system allows the kinetics of editing to be visualised in real time and allows simple scoring of the proportion of cells which have been edited by NHEJ or corrected by HDR. It therefore provides a simple, fast and scalable manner to optimise reagents and protocols for gene correction by CRISPR/Cas9, especially compared to sequencing approaches, and will prove broadly useful to many researchers in the field. Further to this, I have shown that methods which lead to gene correction in our reporter system are also able to partially repair mutations found in the disease-causing gene, Zmynd10; which is implicated in the respiratory disorder primary ciliary dyskinesia (PCD), for which there is no effective treatment. PCD is an autosomal-recessive rare disorder affecting motile cilia (MIM:244400), which results in impaired mucociliary clearance leading to neonatal respiratory distress and recurrent airway infections, often progressing to lung failure. Clinically, PCD is a chronic airway disease, similar to CF, with progressive deterioration of lung function and lower airway bacterial colonization. However, unlike CF which is monogenic, over 40 genes are known to cause PCD. The high genetic heterogeneity of this rare disease makes it well suited to such a genome editing strategy, which can be tailored for the correction of any mutated locus.

Monoclonal antibodies (mAbs), which represent the largest market of any protein therapeutic, are used in the treatment of conditions such as cancer and autoimmune diseases, and prophylaxis of infections. The complexity of manufacture of mAbs translates to a high cost compared with small-molecule drugs, and their use may be restricted as a result. Gene transfer for direct production of mAbs in vivo could offer an alternative to parenteral administration, with the added benefit of easing patient burden caused by repeated injections or infusions. Studies presented in this thesis investigated the possibility of using gene transfer agents for expression of mAbs in the lung lumen and circulation. Using the secreted reporter protein Gaussia luciferase, delivery of the lentiviral vector recombinant simian immunodeficiency virus, pseudotyped with F and HN proteins from Sendai virus (rSIV.F/HN) for targeting of pulmonary epithelium, was identified as an efficient means of gene transfer to the murine lung, with transgene expression in the lung lumen lasting for at least 12 months in BALB/c mice. The recombinant adeno-associated viral vector serotype 8 (rAAV2/8) injected into the muscle directed robust expression of the reporter protein into the circulation and lung lumen of mice for at least 12 months. Two model applications were used to assess the effectiveness of these vectors for mAb gene transfer: respiratory syncytial virus (RSV) infection and rheumatoid arthritis. Both rSIV.F/HN and rAAV2/8 vectors expressing the anti-RSV mAb palivizumab protected mice from weight loss upon RSV infection. Recombinant AAV2/8 delivered intramuscularly was also used to express the anti-tumour necrosis factor alpha (TNFa) biologics infliximab and etanercept for at least 6 months and 56 days, respectively. Due to the immunosuppressive nature of TNFα antagonists, constitutive expression is undesirable and small molecule-assisted shutoff (SMASh) was investigated as a possible method of regulation of etanercept production. In conclusion, viral vectormediated mAb gene transfer to the muscle and lung could be a feasible alternative to parenteral administration provided that an effective method of regulation of expression following vector delivery can be developed.

The bestrophinopathies are a set of inherited retinal degenerations caused by mutations in BEST1, and include Best vitelliform macular dystrophy (BVMD), autosomal dominant vitreoretinochoroidopathy (ADVIRC) and autosomal recessive bestrophinopathy (ARB). The corresponding protein, bestrophin-1, is localised to the basolateral membrane of the retinal pigment epithelium (RPE), where it is thought to function as a Ca2+-activated Cl- channel. Currently, there are no treatments for these conditions. In recent years, gene therapy has emerged as an exciting treatment option for inherited retinal disorders (IRDs). Gene delivery to retinal cells using a recombinant adeno-associated virus (rAAV) has produced positive results in several IRDs. Given the recessive nature of ARB, this thesis proposes that the rAAV-mediated delivery of bestrophin-1 to the RPE could represent a potential therapy. The aims of this thesis were to produce and compare rAAV vectors in vitro and in vivo for protein expression, localisation following transduction, restoration of chloride conductance in vitro and safety following sub-retinal injection in vivo. Following the production of two rAAV vectors expressing bestrophin-1, western blots confirmed bestrophin-1 protein expression following transduction of HEK293 cells in vitro. Immunocytochemistry (ICC) revealed bestrophin-1 expression that was localised to the cytosol. Whole-cell patch-clamping revealed a significant increase in chloride conductance in HEK293 cells transduced with AAV-BEST1 vectors which was then ablated upon the removal of chloride from the buffers. Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS) indicated that the bestrophin-1 protein was successfully transcribed and translated from the BEST1 coding sequence (CDS). Sub-retinal injections of AAV-BEST1 produced bestrophin-1 expression in the RPE of wild-type C57BL/6 mice however significant retinal thinning was seen at higher doses of vector. In conclusion, rAAV-mediated transfer of bestrophin-1 to the RPE has potential to be a future therapy for ARB, however safety issues need to be addressed and an RPE-specific promoter could be more suitable.

Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disease particularly characterised by degeneration of motor neurons from the ventral horn of the spinal cord. Such degeneration leads to muscle weakness and wasting (atrophy). Survival motor neuron (SMN) 1 gene is the SMA-determining gene, being absent or mutated in most people affected. SMN2 is a highly similar gene which produces low levels of SMN protein and allows survival in the absence of SMN1. Full-length SMN is a ubiquitous and essential cellular protein, and low levels of it result in a wide range of systemic pathologies in affected individuals. One avenue to treat SMA is gene therapy, a technology that uses genetic material to alter gene expression and can be applied to both inherited and acquired disorders. Gene addition therapy to replace the faulty SMN1 gene and thus increase the levels of SMN protein to a therapeutic threshold is one of the main strategies for treating SMA. Adeno-associated virus (AAV) vectors are currently being used in clinical trials to attempt such enhancement of SMN levels. A possible alternative may be provided by lentiviral vectors, which are widely used in basic research and clinical applications for gene transfer, including long-term expression. A previous study in Prof. Yáñez's lab demonstrated that lentiviral vectors were significantly more effective for in utero transduction of motor neurons than AAV. Therefore, this study is focused on optimising a lentiviral expression system for SMN1 expression. A variety of integration-proficient and integration-deficient lentiviral vectors (IPLV and IDLV, respectively) with different promoters -either cytomegalovirus (CMV) or human synapsin (hSYN)- and transgenes -either wild-type or a novel codon optimised hSMN1- were produced to compare the efficiency of transgene expression and thus determine the lentiviral configuration resulting in the highest production of full-length SMN. Vectors were tested in a variety of proliferating and quiescent in vitro cell culture models for quantitative estimation of full-length SMN production. The results of these experiments revealed that using integrating, CMV-driven, codon-optimised hSMN1 lentivector resulted in highest production of full-length SMN protein. In general, a comparison of expression levels revealed that IPLVs produced more transgenic protein than IDLVs. However, IDLVs achieved significant transgene expression levels, particularly in quiescent or growth-arrested cells, and they are expected to be significantly safer than IPLVs. Of note, overexpression by hSMN1 lentivectors restored gems in transduced type I SMA fibroblasts in a dose-dependent manner. Encouraged by the results of the cell culture work, we performed very preliminary in vivo experiments. As SMN-replacement therapy is particularly effective at early stages of post-natal development, we hypothesised that earlier treatment may be advantageous. We thus attempted to deliver control and our novel SMN1 lentivectors to the developing embryo of wild-type and SMA mice. While it is clear that this part of the work is at a very early phase, the outcome showed encouraging viability and lack of toxicity in treated animals. In conclusion, the results of this study suggest that lentiviral vectors are potent agents for transgenic expression of codon-optimised hSMN1, and that in utero delivery has significant potential as a therapeutic strategy for SMA.

Spinal muscular atrophy (SMA) is a severe genetic neuromuscular disorder characterised by lower motor neuron degeneration and paralysis. Although it is a leading genetic cause of childhood death no approved treatment options currently exist. As SMA is caused by low levels of the survival motor neuron (SMN) protein the majority of therapeutic strategies under development are therefore aimed at trying to elevate SMN levels. However, a number of limitations with these approaches exist demonstrating a need for the investigation of SMN-independent therapeutics. Of these non-classical pathways, the ubiquitin proteasome system (UPS) is an exciting new area of SMA research. The UPS is a system which degrades unwanted or damaged proteins and alterations in the UPS (including reduced levels of ubiquitin-like modifier activating enzyme 1 [Uba1] and increased levels of ubiquitin carboxyl-terminal esterase L1 [Uchl1] and β-catenin) have been recently identified in the neuromuscular system of SMA mice, providing promising new targets for therapy development. In this thesis I demonstrate that UPS perturbations are also present in other organ systems of severe ‘Taiwanese’ SMA mice and in other SMA models including intermediate Smn2B/− mice, zebrafish and patient derived iPSC motor neurons. Given the previously demonstrated improved neuromuscular phenotype in SMA mice treated with the β-catenin inhibitor quercetin I have been establishing whether other compounds with β-catenin inhibition offer similar or even better therapeutic options. Aspirin, indomethacin and iCRT-14 trials did not improve the SMA phenotype with likely off-target adverse effects meaning that quercetin remains the most tolerable β- catenin inhibitor in SMA mice to date. Another potential target of the UPS for SMA therapeutics is the deubiquitinating enzyme Uchl1, levels of which are increased in SMA. In this thesis I show that pharmacological inhibition of Uchl1 did not improve survival or motor performance in SMA mice and instead had a detrimental impact on the disease phenotype which could be explained by worsening SMA ubiquitin defects. Histological analysis revealed that there was no improvement in lower motor neuron count numbers, neuromuscular junction deficits or muscle fibre diameters. Mimicking the UPS phenotype in primary neuronal cells suggested that targeting UPS perturbations observed in SMA that are upstream of Uchl1, particularly the loss of Uba1, may therefore offer a more effective therapeutic option. Finally, I therefore examined whether increasing Uba1 levels in SMA mice using gene therapy technology was able to improve the SMA phenotype. My initial studies indicate that delivery of AAV9-UBA1 to SMA mice may be beneficial as intraperitoneal injection of AAV9-UBA1 was found to increase the weight and improve motor performance of SMA mice. Intravenous delivery of AAV9-UBA1 was found to further improve expression levels and biodistribution of AAV9-UBA1 in the central nervous system as well as systemically in all body organs and tissues. Western blot and proteomic analysis revealed that AAV9-UBA1 gene therapy is also able to correct downstream UPS perturbations found in SMA as well as increase SMN levels. Together, these results suggest that AAV9-UBA1 gene therapy is an exciting novel therapeutic approach for SMA.

Gene therapy is the introduction or alteration of genetic material with the intention to treat disease. To support this aim, viruses have been modified, with elements linked to viral pathogenicity removed from their genome and replaced by the genetic material to be delivered. Gene therapy vectors based on lentiviruses have many advantages, such as the ability to transduce non-dividing cells and to target specific cell types via pseudotyping. They have been successfully used in ex vivo clinical trials for several haematopoietic stem cell disorders. Lentiviral vectors, however, suffer from substantially lower titres than the more popular adeno-associated virus (AAV)-based vectors and therefore have limited applicability for in vivo gene therapy which requires much greater quantities of virus. The main aim of this thesis was to investigate strategies to improve lentiviral vector productivity during manufacture, in order to increase the likelihood of lentiviruses being adopted for disease treatment. Initial experiments were based on the lentiviral vector manufacturing process currently being developed by the United Kingdom Cystic Fibrosis Gene Therapy Consortium for the generation of highly concentrated, purified lentivirus for clinical use. Supplementation of FreeStyle 293 Expression Medium used during upstream processing was attempted, but none of the assessed supplements led to significant increases in lentiviral vector production. Investigation into intrinsic immunity to viral infection indicated that over-expression of the protein kinase RNA-activated (PKR) led to lower production titres, but over-expression of its inhibitors was not successful at increasing titres. The focus then shifted to reducing, or 'knocking-down', inhibitory factors present in the host cells, which could adversely affect viral titres. Investigation of the published HIV-1 literature revealed a possible 152 candidate inhibitory factors described as having a negative impact on HIV-1 replication in the late stages of the life cycle of the virus. A novel siRNA screen was developed to assess the effect of ‘knock-down' of inhibitory factors on lentiviral vector titre. Application of the screen to 89 candidate inhibitory factors identified nine genes which, when knocked-down, resulted in increased lentiviral vector production by more than 40%. Further work will be necessary to understand the role of the inhibitory factors in lentiviral vector production, but novel cell lines in which genes encoding these factors have been permanently deleted from producer cells could lead to higher titres, reducing costs in the manufacture of lentiviral vectors and making in vivo gene therapy more feasible from a health economics perspective.

Huntington's disease (HD) can be regarded as a model neurodegenerative disorder to screen potential genetic based therapeutics and their carriers. It is an autosomal dominant disorder caused by a mutation in a single gene that leads to progressive neurodegeneration caused in part by protein misfolding. The mutation codes for an expanded polyglutamine tract within the Huntingtin gene (HTT) which leads to neuronal loss through a pathological cascade of events. Current treatment strategies include symptom management but no disease-modifying therapies exist. Research has shown that nucleic acid based therapeutics aimed at decreasing HTT expression can prevent, or reverse, the phenotype. Translating such therapies to the clinic is hindered by the blood brain barrier (BBB) and the lack of an easily administrable, non-toxic, immunologically inert delivery vehicle capable of bypassing the BBB. This study examines a range of nucleic acid based therapeutics for their potency and toxicity, and evaluates extracellular vesicles (EVs) as a delivery vehicle through investigation of the biodistribution of brain targeted EVs and an analysis of EV loading. A small interfering RNA (siRNA) targeting a region upstream of the repeat induced potent non-toxic silencing of HTT. In examining EVs as a carrier for therapeutics for neurological diseases including HD, it was found that targeting can increase brain accumulation of EVs but that the physiological characteristics of EVs which make them susceptible to clearance by the reticuloendothelial system (RES) must be further evaluated to bioengineer modified EVs to avoid fast clearance. Lastly, loading by electroporation was found to induce siRNA aggregation which can lead to overestimation of loading by increasing siRNA content in an ultracentrifugation pellet, but cholesterol-conjugated siRNA mixed with EVs was capable of generating silencing in vitro. This thesis examined EV based treatment for HD through the selection of a nucleic acid cargo to silence HTT, and examination of EVs as a delivery vehicle via biodistribution and loading studies. If the loading can be optimised and fast clearance avoided, there is promise in the use of EVs as a carrier of siRNA for HD.

Collaboration between academia and industry has been the focus of numerous government reports and initiatives over the past 15 years, and is increasingly recognized as an effective way to capitalize on the UK's world-class research base. However, there is a need to further understand the role of such collaborations in the field of translational medicine, where the path to market is particularly lengthy, expensive, and risky, due to complexities associated with the clinical trial process. This research uses a mixed methods approach to investigate collaboration in translational medicine at the University of Oxford. The project comprises three principal stages. First, a broad understanding of the current landscape of academia-industry collaboration in translational medicine was obtained by administering a questionnaire to academics who had received industry funding. Next, a deeper understanding of the barriers to collaboration was sought through semi-structured interviews with 27 academics. Finally, potential strategies to reduce practical barriers to the collaboration process were investigated through interviews with members of three groups within the university who interact directly with industry: Research Services, Oxford University Innovation, and Business Development. This research constitutes the first empirical study on university-industry collaboration in translational research in the United Kingdom. It contributes to existing theory through the development of a new theoretical framework for the evaluation of barriers in terms of a) the practicalities of the collaboration process, b) the institutional environment and c) presiding cultures. Through these analyses, differences in experiences of barriers to collaboration emerged for clinical and non-clinical researchers. Furthermore, industry was seen as playing a crucial role in the translation of new therapeutics, especially in the funding of research that was perceived as being ‘too risky’ for Research Councils. Thus, reducing barriers to university-industry collaboration was seen as important to the realisation of public benefit from university research. Barriers were seen as being overcome, or avoided, via the formation of relationships between academics and companies at several different levels; while systems exist within the university to facilitate this, awareness and uptake of these systems was poor amongst the study population. Finally, if universities are to deliver impact as a key metric of performance, incentives within the university need to reward academics for commercialisation activities, in addition to publication. Through the suggestion of long and short-term strategies and a detailed analysis of industrial collaboration in this setting, this research has implications for both university and government policy.

Radiotherapy is the use of ionising radiation to induce localised DNA damage to cancerous tissues, leading to cell death and disease control. In order to maximise tumour growth control and to limit damage of the healthy surrounding tissues and the consequent side effects for the patient, molecular determinants of tumour radioresistance are investigated as potential clinical targets. A high-throughput siRNA colony formation assay screen in HeLa cervical carcinoma cells previously published by our laboratory identified modulators of radiosensitivity. From the list CSF1R, EPHB2, GAK and TOPK, were selected and validated. TOPK (T-LAK cell-originated protein kinase, also known as PDZ-binding kinase, PBK) was selected for further investigation because it is overexpressed in most malignancies but not in normal tissues, apart from testis and placenta. Knockdown of TOPK was shown to induce radiosensitisation in a panel of cancer cell lines with no significant effects on normal cells. A role for TOPK in the cell cycle response to ionising radiation (IR) was discovered in HCT116 colorectal cancer cells, with alterations in the G1/S and G2/M checkpoints. Furthermore, immunoprecipitation experiments identified a physical interaction between TOPK and CDKN1A (p21) at 8 hours after IR. Apoptosis and the number of multinucleated cells were significantly increased in TOPK depleted cells exposed to IR, suggesting the possibility of aberrant mitosis and mitotic catastrophe in these cells. High TOPK expression in early breast cancer patients was shown to be associated with poor recurrence-free survival. In addition, immunohistochemistry (IHC) analysis on samples from prostate cancer patients identified a strong correlation between high levels of TOPK and poor clinical response to radiotherapy. In order to facilitate future in vivo experiments, an HCT116 shRNA stable knockdown cell line was developed and two commercially available TOPK inhibitors were tested and optimised. Taken together, these data suggest that TOPK is a molecular determinant of radiosensitivity with a great potential for future clinical applications.

Friedreich ataxia (FRDA) is a progressive neurodegenerative disease with primary sites of pathology in the large sensory neurons of the dorsal root ganglia (DRG) and dentate nucleus of the cerebellum. FRDA is also often accompanied by severe cardiomyopathy and diabetes mellitus. FRDA is caused by loss of frataxin (FXN) expression, which is due to GAA repeat expansion in intron 1 of the FXN gene. Frataxin is a mitochondrial protein important in iron-sulphur cluster (ISC) biogenesis and in the electron transport chain (ETC). As a consequence of impaired mitochondrial energy metabolism, FRDA cells show increased levels of and sensitivity to oxidative stress, which is known to be associated with genome instability. In this study, we investigated DNA damage/repair in relation to FXN expression via immunostaining of γ-H2AX, a nuclear protein that is recruited to DNA double strand breaks (DSBs). We found FRDA patient and YG8sR FRDA mouse model fibroblasts to have inherently elevated DNA DSBs (1.8 and 0.9 foci/nucleus) compared to normal fibroblasts (0.6 and 0.2 foci/nucleus, in each case P < 0.001). By delivering the FXN gene to these cells with a lentivirus vector (LV) at a copy number of ~1/cell, FXN mRNA levels reached 48 fold (patient cells) and 42 fold (YG8sR cells) and protein levels reached 20 fold (patient cells) and 3.5 fold (YG8sR cells) that of untreated fibroblasts, without observable cytotoxicity. This resulted in a reduction in DNA DSB foci to 0.7 and 0.43 (in each case P < 0.001) in human and YG8sR fibroblasts, respectively and an increase in cell survival to that found for normal fibroblasts. We next irradiated the FRDA fibroblasts (2Gy) and measured their DSB repair profiles. Both human and mouse FRDA fibroblasts were unable to repair damaged DNA. However, repair returned to near normal levels following LV FXN gene transfer. Our data suggest frataxin may be important for genome stability and cell survival by ensuring ISC for DNA damage repair enzymes or may be required directly for DNA DSB repair.

Cystic Fibrosis (CF) is a degenerative disorder that is often associated with chronic lung disease. CF is caused by mutations in the gene encoding the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) chloride channel, which lead to defective chloride and sodium ion movement across epithelia. Subsequent dehydration of the airway surface liquid (ASL) on airway epithelia, is associated with poor mucociliary clearance and chronic lung infection. The monogenic nature of CF, along with the accessibility of the lung, makes the disease amenable to gene replacement therapy. Gene therapy clinical trials have focused on replacing the mutated CFTR with a functional copy, which has led to improved chloride transport, but has shown no significant effects on sodium transport. An alternative strategy for CF gene therapy therefore, could be to reduce the expression of the epithelial sodium channel (ENaC) in the lung, using RNA interference (RNAi), combined with CFTR delivery. Developing a dual-function gene transfer vector could potentially restore chloride and sodium levels in the ASL and help alleviate CF lung disease. The aim of this thesis was to develop a recombinant lentivirus delivery system capable of simultaneously delivering CFTR expression and knocking down ENaC expression in the airways. A modular HIV vector genome plasmid was developed to allow simple insertion of various promoter elements, transgenes and knockdown sequences, for subsequent virus production. Insertion of the CFTR transgene and a short-hairpin RNA (shRNA) sequence targeting the ENaC alpha subunit (ENaCα) resulted in significant knockdown of human ENaCα and simultaneous expression of CFTR in A549 (human lung carcinoma) cell culture. Replacement of the ENaCα shRNA with an shRNA targeting the transcription factor BACH1 resulted in target gene knockdown and concomitant HMOX1 up-regulation, confirming specific knockdown effects, and demonstrating that the dual-function rLV vector could mediate target gene knockdown irrespective of the target. Attempts were made to knock down BACH1 in primary cultures of human bronchial epithelial cells grown at the air-liquid interface (ALI), but improved transduction efficiencies from the apical surface will be required to generate successful knockdown in this experimental model. These studies provide proof-of-principle for the utility of this versatile dual-function prototype virus. The dual function vector not only has the potential for treatment of CF lung disease, but could be readily altered to target other lung diseases where combinations of prolonged target gene knockdown and gene expression/up-regulation could collectively provide an appropriate therapy. In conclusion, the focus on the rational design of gene transfer vectors for specific therapeutic effects will aid the development and translation of gene therapy approaches to human studies.

Despite expressing antigens that can induce immune surveillance and immune eradication, tumours demonstrate the capacity to evade anti-tumour immunity. Recently, this has been attributed to the ability of tumours to induce a local immunosuppressed micro-environment, which is a major obstacle to successful natural and vaccine induced anti-tumour immunity. Soluble factors such as transforming growth factor beta (TGFβ), and interleukin 10 (IL-10), released by cancer and stromal cells, are thought to play a significant role in this local immunosuppression. In order to assess the influence of antagonising these soluble factors locally on tumour biology and tumour immunity, a murine CT26 colorectal carcinoma model that can express cytokine antagonists under Doxycycline (Dox) control was engineered. Two stable CT26 cell lines expressing Dox-inducible soluble extracellular domain of TGFβ receptor II (STGFβRII) or soluble extracellular domain of IL-10 receptor (SIL-10R), were established. Expression of STGFβRII in vitro and in vivo was only evident after Dox treatment. When Dox was administered directly following subcutaneous (s.c.) inoculation of STGFβRII-expressing CT26 cells into Balb/c mice, tumour growth was significantly suppressed. Interestingly, inducing STGFβRII in well-established tumours showed less suppression of tumour growth. To assess the effect of expressing STGFβRII on tumour immunity the RNA expression of 22 common cytokines was measured in the tumours of mice receiving Dox and control mice. The levels of some of these cytokines were modulated by STGFβRII expression (e.g TGFβ,Tnfsf9, IL-2). We also tested for any additive effect between expressing STGFβRII, and the administration of anti-CTLA-4 antibody on tumour growth, and tumour immunity. The model described here could help address various limitations seen previously in studies of TGFβ blockade. It provides means of effective local antagonism, and it addresses immunological endpoints which have been limited in previous studies. Because of its tight regulation, the model also allows identification of the best timing of TGFβ blockade alone or in combination with other immunotherapeutics.

Next-generation sequencing technology has transformed all areas of molecular biology and fueled a dramatic increase in the availability of viral genomic sequences. Assessing the functional properties of viral genes and genetic elements, however, remains a major bottleneck, and linear investigation of single candidates of interest is increasingly insufficient to keep pace with the demands of basic and translational virology. Assays that parallelize the evaluation of thousands or millions of viral sequences offer an opportunity to scale up characterization of both naturally occurring viruses and engineered viruses with therapeutic potential for gene delivery. In this work I develop and apply high-throughput screens for two purposes: the development of engineered vectors for gene therapy and the identification and characterization of cis-regulatory translation control strategies across naturally occurring viruses. Recombinant adeno-associated virus (AAV) vectors are promising vehicles for gene delivery to treat a wide variety of genetic disorders through gene replacement, gene editing, or gene silencing, but naturally occurring AAV capsids present significant limitations for therapeutic applications. Engineered AAV capsids can expand the repertoire of available vectors possessing desirable properties with respect to tissue tropism and off-target effects. High-throughput approaches to capsid engineering that harness the power of next-generation sequencing to maximize the search space of possible capsid variants hold great promise for this end. In Chapters 2 and 3, I present DELIVER (Directed Evolution of AAV capsids Leveraging In Vivo Expression of transgene RNA), a massively parallel approach to capsid engineering that yields capsids with an enhanced ability to express a transgene in a particular tissue and cell type. In Chapter 2, I describe the properties of the MyoAAV class of capsids that were developed using DELIVER and potently transduce skeletal and cardiac muscle cells of both mice and juvenile macaques. These capsids have immediate potential for clinical applications in the treatment of degenerative musculoskeletal disorders with a genetic basis. I also identify the interaction between MyoAAV and the integrin aVb6 heterodimer as the key determinant of MyoAAV's muscle-specific tropism. In Chapter 3, I apply DELIVER to develop the PAL class of AAV capsids that demonstrate enhanced transduction of the central nervous system (CNS) and reduced off-target transduction of the liver in juvenile macaques. The PAL capsids are among the first reported engineered capsids to show an enhanced ability to transduce the primate CNS compared to naturally occurring capsids. In Chapter 4, I apply a high-throughput technique to a more exploratory problem in molecular biology by using a massively parallel reporter assay to investigate upstream open reading frames (uORFs) in viruses. uORFs—short ORFs initiating 5' of a main coding sequence (mCDS)—are well-known cis-acting negative regulators of translation in eukaryotes. In many cases, uORFs can dynamically respond to changing intracellular conditions and derepress the downstream mCDS in response to specific signals. Despite their importance in key pathways in eukaryotes, uORFs remain understudied in viruses. Our massively parallel approach facilitates systematic characterization of regulatory sequences from hundreds of human-associated viruses, a feat that would be intractable through more traditional experimental means. I provide quantitative evidence that regulatory uORFs are common across a broad range of viruses, are actively translated by ribosomes, and share some fundamental properties with eukaryotic uORFs. Taken together, this dissertation presents a multi-pronged exploration of the utility of massively parallel approaches in the study and translational application of virology.

Viral vectors have become crucial to multiple fields as vehicles for gene therapy, vaccines, and reporter genes, among other uses. Two of the most prominent vectors were generated from adenoviruses and adeno-associated viruses (AAVs). Presently, adenoviral vectors have been clinically studied as vaccines for Zika, Ebola, and SARS-CoV-2 viruses. AAVs are now used in FDA approved gene therapy treatments. While vectors have already had an immense impact on the scientific and medical world, there remain many possible new applications as well as areas of improvement for these tools. In chapter two a new virome (SIV infected rhesus macaque) is explored allowing for the discovery of five novel AAV capsids. Structure guided analysis of the RhAAV4282 capsid to the closely related AAV7 capsid enabled the identification of a novel AAV capsid neutralization epitope named the Hypervariable Region IV Neutralizing Epitope (HRNE). Swapping the HRNE between RhAAV4282 and AAV7 capsids significantly reduced neutralizing antibody titers against the parental capsids. In chapter three the adenovirus serotype 26 (Ad26) vector platform is used to deliver neoantigen vaccines in a cancer model. Ad26 and long peptide neoantigen delivery were analyzed by traditional mechanistic immunology methods and scRNA-seq/TCR-seq. This analysis showed increased immune responses, upregulation of maturation and activation T cell pathways, clonal hyper-expansion, and protection by the Ad26 platform but no difference in tumor T cell infiltration. These results highlighted the importance of not just T cell quantity, but also T cell quality induced by a cancer vaccine. In chapter four the proteomes of 82 different sarbecoviruses were aligned to identify conserved regions, with the goal of generating a broadly protective sarbecovirus vaccine. A portion of the sarbecovirus RNA-dependent RNA polymerase and helicase were identified as highly conserved (93% perfect T cell epitope conservation) among sarbecoviruses. A rhesus adenovirus serotype 52 (RhAd52) vaccine carrying this immunogen was found to be immunogenic and protective against challenge by mouse adapted SARS-CoV-2. The data presented in this thesis generates new possibilities for AAV capsid engineering, furthers understanding of how to design neoantigen cancer vaccines, and identifies a highly conserved sarbecovirus immunogen. This research has implications for advances in viral vector applications, vaccines, cancer, and gene therapies.

For many dominant diseases, conventional treatment options are limited. This makes them attractive candidates for gene therapy, which may be directed to specifically silence a disease-causing allele. However, many mutations are not easily amenable to this technique, including nucleotide repeat expansions, which cause numerous neurodegenerative diseases such as Huntington’s disease and several Spinocerebellar Ataxias. Combined gene knockdown and replacement (K&R) may present a more practical approach for such conditions, whereby the gene of interest is subject to mutation-independent non-allele-specific silencing and concurrently replaced with a functional copy. Artificial mimics of naturally occurring intronic microRNAs are theoretically ideal for this purpose, since they can be nested within the replacement gene. This thesis investigates the development of mimics of two different intronic miRNA systems (mirtron miR-1224 and the miR-106b cluster) to silence Ataxin-7 and to be incorporated into novel single K&R constructs for testing in vitro. Artificial mirtrons and intronic miRNAs were successfully developed and shown to silence Ataxin-7 mRNA in numerous cell lines. An RNAi-resistant gene was developed and mirtrons could be successfully incorporated as introns. Patient-derived fibroblasts and iPSC-derived neuronal cells were investigated as models for testing of gene silencing therapies. This work suggests that mirtron-based K&R is achievable, and warrants further investigation.

Retinitis pigmentosa (RP) is a genetically and phenotypically heterogeneous condition that affects approximately 1 in 4000 individuals worldwide. The most common presentation of RP is a rod-cone dystrophy, where the degeneration of cone photoreceptors occurs secondary to advanced rod loss, leading to a significant decline in central vision and a corresponding reduction in patient quality of life. The mechanisms underlying secondary cone loss are poorly understood, particularly in disorders where the gene defect is unknown or manifest only in rod photoreceptors. Consequently, the thesis presented herein proceeds on several fronts. First, in the long term a greater understanding of the causes underlying cone loss in RP is likely to be beneficial, and so in chapter one a dominant cone degeneration is characterized using intrinsically fluorescent cone photoreceptors to track the degenerative process. Second, as we develop a greater understanding of the genetic etiology underlying RP it is likely that the number of large genes identified as being causative will increase. As currently there is no efficient way to deliver large genes to photoreceptors, chapter two explores the use of alternate viral vectors that might be used to deliver a large therapeutic transgene. Lastly, whilst our understanding of cone loss in RP remains incomplete, it is necessary to develop a broadly applicable therapy to slow or attenuate further cone loss in RP patients regardless of the underlying cause. In chapters three and four we examine the use of low molecular weight "growth factors‟, such as ciliary neurotrophic factor (CNTF), to preserve cone photoreceptors long-term using a rhodopsin knockout model of RP.

Retrovirus gene therapy vectors can deliver therapeutic genes to mammalian cells in a permanent manner by integrating their genome into host chromosomes and therefore provide the potential for long term therapeutic gene expression. Retrovirus integration, however, can be oncogenic. Apart from insertional mutagenesis (IM) genotoxicity may be caused by other factors including DNA damage following infection and integration and epigenetic effects related to incoming viral particles. Thus, using retrovirus and lentivirus infected murine tumour tissue and infected cell lines in vitro this thesis was directed at investigating whether virus infection and integration could cause genotoxicity by alternative route(s) other than IM. Using clonally derived liver tumours that developed in mice, and normal liver and kidney tissues, following EIAV and HIV delivery in utero, comparative genome hybridisation methodology was used to examine for copy number variation. This showed amplification and deletions only in EIAV derived tumours. Real time Q-PCR analysis was then used to measure gene expression changes relating to genes contained within or near to amplifications observed in two tumours of individual mice. The STRING database was then used to find networks linking genes with differential expression profiles and genes in one of these tumours identified with provirus insertions that were also differentially expressed. These data provided preliminary data implicating a role for LV in Hepatocellular carcinoma (HCC). DNA damage is known to cause chromosomal instability that can lead to tumour development. The relationship between double strand breaks (DSB) and virus infection was also investigated in-vitro to find alternative routes to genotoxicity other than IM. Cell viability analysis demonstrated cells with a defective DNA damage response (DDR) have decreased cell viability compared with cells with intact DDR when infected with RV or LV vectors. DSB assays showed RV and LV infection to generate foci over a 6 hour period followed by DDR. Where no viral integrase is present, no DDR appears, however, where the vector is used with or without a genome to infect cells, DDR occurs as shown by the presence of 53BP1 foci indicative of DNA damage. The relationship between DNA damage and methylation was also investigated. Global methylation was found elevated in the genomic DNA of LV and RV infected cells and not in control uninfected cells. In contrast, methylation changes were not found in infected cells lacking the NHEJ repair pathway. These data suggest the DNA damage response is linked to genome methylation. The E2F transcription factor plays a key role in regulating expression of genes known to control oncogenesis and cancer, and E2F is regulated by methylation of its related target gene promoters. Taking into account all genes in the human genome the number of genes that bind E2F is 32.77%. However, using microarray to represent genes differentially expressed after infection, 59% of these were E2F targets. Overall, taking the data obtained in this thesis into account it may be suggested that RV and LV infection causes a number of potentially related changes to cells that include DNA damage and repair and methylation changes that could influence E2F that is an important factor involved in oncogenesis. Combining this with IM, attenuated RV and LV currently in use for gene therapy may cause genotoxicity to infected cells and increase the risk of oncogenesis especially where DNA damage is not correctly repaired. Further work is required to show in greater detail the extent of this genotoxicity, possible by whole genome sequencing of treated host genomes or cell transformation assays linked to the genotoxicity assays presented here. Collectively these data show that alternative factors to IM might exist that could act independently or synergistically to IM.

Gene therapy consists of methods which attempt to repair or replace defective genes responsible for disease, or to add genes to a therapeutic effect. To achieve this, two episomally maintained recombinant viral vectors have shown promising results: integration-deficient lentiviral vectors (IDLVs), and adeno-associated virus (AAV) vectors. The non-integrating nature of these vectors improves their safety profile but also limits transgene retention as nuclear episomes generally get lost during cell division. In the present study, the establishment of stable replicating episomes via transduction with AAV and IDLV gene therapy vectors was examined in CHO cells. Different DNA elements and cell culture conditions were evaluated, and in particular the effects of (i) DNA elements called S/MARs (scaffold/matrix attachment regions) which are involved in chromatin organisation, transcription and replication, and (ii) induction of transient cell cycle arrest in transfected and transduced cell populations. In the case of both AAV and IDLV vectors, the incorporation of S/MAR elements into vector transcription units had only marginal effects on the establishment of stable transgene-positive cell populations, either with or without induction of transient cell cycle arrest. However, a striking general result was observed in cell populations transduced with IDLVs and subjected to a transient cell cycle arrest soon after transduction. Under these conditions, following release from cell cycle arrest and in the absence of any selection pressure, substantial populations (10-25%) of proliferating and stably transduced cells emerged and were maintained over at least 100 population doublings. This establishment of stable transduction was seen only with IDLVs, was crucially dependent on the induction of a period of transient cell cycle arrest, occurred independently of the presence of S/MAR elements, and resulted in transgene-positive cell populations which could be isolated and propagated as stable clonal cell lines. In these polyclonal and clonal IDLV-transduced cell lines, the existence of non-integrated vector genomes in the form of multi-copy nuclear episomes was confirmed by evidence from linear amplification -mediated PCR, deep sequencing, Southern blotting and FISH (fluorescent in situ hybridisation). The cumulative evidence suggests that transduction of CHO cells with IDLVs followed by a short period of induced cell cycle arrest leads to the establishment of stable IDLV-based nuclear episomes which are transcriptionally active and undergo replication and segregation during cell division without the need for antibiotic-based or other positive selection pressure. Preliminary investigations were also done to test the capacity of combined IDLV transduction and transient cell cycle arrest to establish stable episome HeLa cells and murine haematopoietic stem cells. However, further experiments are required either to optimise the protocol in these cells or to find other clinically relevant cell types in which the protocol can be implemented. The transfer of this technology to a variety of clinically relevant human stem or progenitor cell populations could improve the safety profile of a range of gene therapy strategies currently under investigation.