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.

Several tools were developed to help researchers facilitate clinical translation of the use of engineered nucleases towards their disease gene of interest. Two major issues addressed were the inability to accurately predict nuclease off-target sites by user-friendly \textit{in silico} methods and the lack of a high-throughput, sensitive measurement of gene editing activity at endogenous loci. These objectives were accomplished by the completion of the following specific aims. An online search interface to allow exhaustive searching of a genome for potential nuclease off-target sites was implemented. Previously discovered off-target sites were collated and ranking algorithms developed that preferentially score validated off-target sites higher than other predictions. HEK-293T cells transfected with newly developed TALENs and ZFNs targeting the beta-globin gene were analyzed at the off-target sites predicted by the tool. Many samples of genomic DNA from cells treated with different ZFNs and TALENs were analyzed for off-target effects to generate a greatly expanded training set of bona fide off-target sites. Modifications to the off-target prediction algorithm parameters were evaluated for improvement through Precision-Recall analysis and several other metrics. An analysis pipeline was developed to process SMRT reads to simultaneously measure the rates of different DNA repair mechanisms by directly examining the DNA sequences. K562 cells were transfected with different types of nucleases and donor repair templates in order to optimize conditions for repairing the beta-globin gene. This work will have significant impact on future studies as the methods developed herein allow other laboratories to optimize nuclease-based therapies for single gene disorders.

published_or_final_version / abstract / Anatomy / Doctoral / Doctor of Philosophy

published_or_final_version / abstract / Surgery / Doctoral / Doctor of Philosophy

published_or_final_version / Medicine / Doctoral / Doctor of Philosophy

Vaults, recently discovered in 1986, are multi-subunit organelles with a molecular mass of ,--,13 MDa. The specific function of vaults is unknown, although they are believed to be involved in internal transport. These ribonucleoproteins are composed of the major vault protein, which comprises ' 70% of the vault's mass, two minor proteins, TEP1 and vPARP, and untranslated RNA(s). It is believed that the protein components of the vault are structural while the RNAs are the functional components. Implications of the vault's involvement in multi-drug resistance in cancer have been made. In some resistant cancer cells, the major vault protein and vRNA(s) are up-regulated up to 15 times when cells are exposed to a cytotoxic drug. Cytotoxic drugs such as doxorubicin are administered as a cancer treatment, but may be ineffective because the drug is actively pumped out of the cell. Multi-drug resistance is the most common failure of chemotherapeutic cancer treatment. In order to prevent the development of multi-drug resistance this research employed the use of small interfering RNA technology to down-regulate the expression of one of the vault RNAs, vRNA3, in cultured GLC4 cells, a small-cell lung cancer cell line. If the vRNA(s) are the functional portion of the vault and a cloned siRNA prevents their up-regulation after drug exposure, the cells should lose their multi-drug resistance, stimulating apoptosis. If successful, this approach may provide an alternative approach to cancer treatment in cells which respond to chemotherapy by increasing the number of vault particles.Initially, the transfection of a plasmid into GLC4 cells was optimized. The best transfection efficiency (N20%) was obtained by using GeneTherapySystems' GenePORTER2 transfection reagent in serum free conditions. To determine if the vault RNAs are the functional portion of the vault complex that confers multi-drug resistance to a cell, a small interfering RNA fragment was designed to specifically knock-down the expression of human vault RNA 3. The siRNA sequence homologous to a portion of vault RNA3 was cloned into an expression vector, and using optimized transfection protocols was transfected into GLC4/REV cells. A Western analysis using caspase-8 antibodies showed no difference in caspase-8 expression in doxorubicin treated and untreated cells. Preliminary results yielded by reverse transcriptase polymerase chain reaction amplification of isolated RNA indicated that the vRNAs were not down-regulated by the siRNAs. / Department of Biology

This thesis describes the design of a virotherapy strategy capable of destroying both breast cancer vasculature and tumour cells, using an oncolytic adenovirus expressing angiogenesis-regulating proteins. Five oncolytic adenoviruses were compared to identify the best virotherapy agent for breast cancer, including measurement of cytotoxicity in vitro, and replication, intra-tumoural spread and anticancer efficacy in vivo. The viruses tested were Ad-dl922-947 (targets G1-S checkpoint defects); Ad-Onyx-015 and Ad-Onyx-017 (target p53/mRNA nuclear export defects); Ad-vKH1 (targets Wnt pathway defects) and AdEHE2F (targets estrogen receptor/G1-S checkpoint/hypoxia signalling defects). AdEHE2F demonstrated optimal oncolytic activity and selectivity against breast cancer, accordingly this virus was engineered to express potent regulatory angiogenic proteins, namely soluble Flt1 and soluble Delta like-4 (Dll4). sFlt1 is the soluble extra-cellular domain of VEGFR1 and binds to and sequesters VEGF-A, thereby preventing VEGFR2 stimulation which is crucial to trigger angiogenesis. sDll4 is the soluble extracellular domain of Dll4 and has been previously shown to block Dll4/Notch signalling. Dll4/Notch signalling increases a chaotic and non-functional angiogenesis which ultimately delays tumour growth. Importantly, VEGF and Dll4 are the only angiogenesis genes reported to be haploinsufficient in vascular development and both have been shown to have a good anti-tumour effect. sFlt1 and sDll4 genes were substituted for the viral genes E3 6.7K/gp19K of AdEHE2F, thereby using endogenous adenoviral machinery to drive production. The activities of AdEHE2F viruses expressing either sFlt1 or sDll4 were compared in vitro and in vivo. sFlt1 (expressed from AdEHE2F) inhibited endothelial cell proliferation and sprouting whereas sDll4 increased proliferation and branching in vitro. In vivo AdEHE2F expressing sFlt1 or sDll4 both showed superior anticancer activity compared to parental AdEHE2F, indicating at least additive efficacy between virotherapy and regulatory angiogenic approaches.

The Wilms' tumor 1 gene is overexpressed in many types of cancer and is associated with poor prognosis and resistance to anti-cancer therapies. In vitro studies in non-glioma cells types have demonstrated that WTl plays a role in increased proliferation, resistance to apoptosis, and increased cellular invasion. We aimed to thoroughly characterize the expression pattern of Wilms' tumor 1 in human malignant glioma and discern its function in this complex disease process. We screened a large sample of established human malignant glioma cell lines and glioma tissue specimens of all grades for WT1 expression. The majority of cell lines and 80% of all glioma tissue expressed WTI mRNA, all of which expressed WTl(+KTS) isoforms. Further screening of the glioblastoma specimens for p53 mutation followed by logistic regression analysis demonstrated a positive correlation between WTl expression and wild-type p53 (p = 0.04). To determine if WTl and p53 functionally interacted, we generated LN-229 glioblastoma cells that stably expressed WTl. As LN-229 cells harbor a p53 mutation, transient transfection with wild-type p53 induced apoptosis. However, stable WTI expression did not protect cells from p53-mediated cell death. We then generated U87MG cells (p53 wild-type) that stably expressed WT1 to model an endogenous p53 response. It is well known that after treatment with ionizing radiation, U87MG cells readily undergo p53-mediated apoptosis. Again, WTI expression did not protect against ionizing radiation induced p53-mediated cell death. We next examined the effect of transient WTI silencing on ionizing radiation induced cell death in T98G and LN-18 cells which express endogenous WTl. Combination treatment with ionizing radiation and silencing of WTI using short interfering RNA caused a decrease in viability and clonogenic survival relative to radiation alone in both cell lines. Lastly, we studied the effect of stable WTl silencing using short hairpin RNA on glioblastoma cell tumorigenicity. Stable transduction of U25 1MG and LN-18 cells with WTI short hairpin RNA resulted in a marked decrease in proliferation. WTI silencing in U251MG cells also caused a decrease in in vitro invasion. WTl silencing in U251MG cells caused an increase in tumor latency and a decrease in tumor growth rate when cells were used to subcutaneously inoculate nude mice. Not only do these studies support an oncogenic role for WTI in glioma biology, they provide encouraging evidence that WTl may be a therapeutic target for molecular treatment of glioblastoma.

The impact of Ad.mda-7 on the survival of renal cell carcinoma lines (RCC), primary renal epithelial cells, glioblastoma multiforme lines (GBM), and primary rodent astrocytes is unknown. The present studies examine whether the GST fusion protein, GST-MDA-7, and the adenovirus, Ad.mda-7, altered the growth and survival of the A498 and UOK121N RCC lines or radiosensitized GBM, respectively. Due to previous findings that the RCC lines, but not primary renal epithelial cells, were resistant to type 5 adenoviral infection, we used purified GST-MDA-7 protein to show that GST-MDA-7, but not GST, caused a dose-dependent reduction in A498 and UOK121N proliferation but not that of primary renal epithelial cells. Free radical species, generated by clinically relevant concentrations of arsenic trioxide, synergized with subnanomolar concentrations of GST-MDA-7 to inhibit the proliferation, viability, and long-term survival of RCC. We also found that MDA-7 (IL-24), when expressed via a recombinant replication defective adenovirus, Ad.mda-7, exerted anti-proliferative effects on GBM cells, an effect found to be enhanced in a greater than additive fashion when combined with ionizing radiation. These findings argue that MDA-7, in combination with agents that generate free radicals, such as arsenic trioxide and ionizing radiation, may have potential in the treatment of RCC and GBM. Geldanamycins are currently being used in a number of clinical trials in different tumor cell types, such as hepatocellular carcinoma (HCC), targeting the inhibition of the heat shock protein and molecular chaperone Hsp90. Previous studies have demonstrated that geldanamycins have dose limiting toxicity in vivo due to their actions in promoting normal liver dysfunction. These studies show that the geldanamycin derivative, 17-allylamino-17-demethoxygeldanamycin (17-AAG), interacts with the secondary bile acid, deoxycholic acid (DCA), to kill primary rat hepatocytes and HuH7 human hepatoma cells. An effect abolished by the addition of the ROS-quenchers, NAC and Trolox. Collectively, these findings argue that geldanamycins may not be a viable therapy for HCC treatment and that 17-AAG toxicity in primary hepatocytes may be, at least upon initial drug exposure, due to ROS generation and mechanisms independent of Hsp9O inhibition and the down-regulation of "classical" Hsp90 client proteins.

Dans la prise en charge de la polyarthrite rhumatoïde (PR), l’identification de biomarqueurs associés à un diagnostic, un pronostic mais surtout à une bonne réponse thérapeutique est un des enjeux de la prochaine décennie, d’autant que de nombreux outils sont maintenant disponibles grâce à l’étude du génome, du transcriptome et du protéome. Au cours de mon travail de thèse, je me suis intéressée à l’identification de différentes classes de biomarqueurs prédictifs de la réponse au traitement par biothérapies.Tout d’abord par une approche protéomique, j’ai identifié une combinaison de biomarqueurs protéiques sériques, MCP-1 et EGF, permettant à l’initiation du traitement par étanercept (anti-TNFα), puis par rituximab (anti-CD20), de prédire la réponse clinique à 3 mois. En parallèle, toujours chez des patients PR traités par rituximab, j’ai recherché un biomarqueur cellulaire par cytométrie de flux. J’ai ainsi pu montrer que la densité de CCR5 à la surface des lymphocytes T CD4+ circulants est faible chez les patients PR . Celle-ci est corrélée positivement à l’activité de la maladie et négativement au taux intracellulaire d’ARNm de CCR5. Trois mois après l’initiation du traitement, la densité de CCR5 à la surface des lymphocytes T CD4+ circulants augmente proportionnellement à la diminution de l’activité de la maladie. J’ai également étudié l’intérêt de biomarqueurs génétiques. Tout d’abord, en analysant le profil d’expression de 723 micro(mi)ARNs dans le sang de patients PR, j’ai montré que l’expression de 37 miARNs était dérégulée par rapport à des donneurs sains. Par ailleurs, j’ai montré que le niveau d’expression de miR-125b dans le sang des patients PR à l’initiation du traitement par rituximab permet de prédire la réponse à 3 mois. J’ai enfin exploré les polymorphismes de gènes de cytokines et de récepteurs aux cytokines afin de prédire la réponse au traitement par rituximab à 6 mois chez 63 patients atteints de PR active. Douze polymorphismes de nucléotides uniques (SNPs) ont été génotypés et ont permis d’identifier 2 SNPs du facteur de croissance transformant beta 1 (TGFb1) codon 25 et codon 10 prédictifs d’une bonne réponse au rituximab.Je me suis aussi intéressée à la thérapie génique qui est un outil thérapeutique pour délivrer des agents anti-TNF dans la PR. En effet le transfert de gène permet à l’organisme de synthétiser in situ la protéine médicament et d’éviter des administrations répétées comme c’est le cas avec les biothérapies. Le défi de demain réside plus dans le développement de vecteurs fiables et efficients. Dans un premier travail réalisé dans l’équipe de recherche du Pr Hirsch à Pittsburgh (E.U), j’ai évalué in vitro l’intérêt d’un virus adéno-associé recombinant de sérotype 2 (rAAV2) contenant un ARN double brin pour le transfert de gène dans les synoviocytes humains. Dans un contexte inflammatoire, celui-ci s’est révélé efficace lorsqu’associé à un inhibiteur du protéasome, le zLLL. Lors d’un deuxième stage effectué au sein du laboratoire du Pr Jorgensen (Montpellier), j’ai évalué le potentiel thérapeutique d’un rAAV de sérotype 5 (rAAV5) exprimant un petit ARN interférant ciblant le TNF-α dans un modèle expérimental d’arthrite. L’inhibition du TNF-α a été validée in vitro sur des macrophages murins, puis in vivo dans le modèle murin d’arthrite au collagène, après injection dans les articulations arthritiques. Depuis la fin des années 90, l’avènement des biothérapies a permis de bouleverser l’évolution de la PR. Les biomarqueurs, en particulier ceux permettant le suivi des traitements, sont des outils de choix à développer pour notre pratique clinique courante afin de choisir d’emblée le traitement le plus efficace et le mieux toléré pour chaque patient. D’autres thérapies innovantes, telles que la thérapie génique, s’appuient sur les succès des biothérapies pour permettre d’autres avancées en choisissant d’emblée les cibles les plus pertinentes. / Rheumatoid arthritis (RA) is a chronic inflammatory disease. The effective use of biologics, which block key molecules involved in the pathogenesis of RA, has dramatically improved the treatment of this chronic disease over recent years. However, for at least one-third of patients with RA biologic treatments will produce an inadequate response. Therefore, the use of predictive biomarkers of response to identify individuals who are likely to respond to specific treatments will provide benefit. An individualised approach will allow patients to receive effective treatment without unnecessary exposure to potentially toxic side effects.During my thesis, I identified predictive biomarkers in RA patients treated with biologics by using proteomics, genetics and cellular biomarkers.Genen therapy could be used to optimise drug administration by avoiding repetitive administration. I validated the interest of a recombinant adeno-associated virus in RA synovial fibroblasts.