Les Flavivirus neuropathogènes, tels que le virus de l’encéphalite japonaise (JEV), le virus West Nile (WNV), le virus de la fièvre jaune (YFV) et le virus Zika (ZIKV) causent des maladies neurologiques. Ces maladies sont dues à une infection des cellules du système nerveux central (CNS) par ces virus. Le CNS est un organe privilégié, isolé des agents pathogènes par une barrière entre le sang et le cerveau, appelée barrière hémato-encéphalique (BBB). Les Flavivirus neuropathogènes capables de traverser cette BBB afin d’atteindre leurs cellules cibles, localisées dans le CNS, sont neuroinvasifs. Le but de cette étude est de comprendre les mécanismes cellulaires permettant aux Flavivirus de traverser la BBB et les effets de l’infection par les virus ZIKV et WNV des cellules du CNS sur le développement de celles-ci.Le YFV est un virus hépatotrope, infectant majoritairement le foie et les reins. Deux vaccins vivants atténués dirigés contre le YFV, le vaccin FNV (pour French Neurotropic Virus) et le vaccin 17D, ont été obtenus empiriquement par passages successifs de souches virulentes de YFV sur cerveaux de souriceaux. Ces vaccins ne causent plus de maladies touchant les reins et le foie, mais peuvent parfois causer des encéphalites post-vaccinales. Ces cas d’encéphalites démontrent que ces souches vaccinales sont devenues neurovirulentes mais aussi neuroinvasives car les virus ont pu franchir la BBB. A cause d’une incidence trop élevée d’encéphalites post-vaccinales par rapport au vaccin 17D, le vaccin FNV a été retiré du marché dans les années 1980.Le JEV est un virus neurotrope, causant des encéphalites graves en Asie du Sud-Est. A ce jour, il existe un vaccin vivant atténué, le JEV SA14-14-2, obtenu empiriquement par passages successifs d’une souche virulente sur cellules de hamster. Ce vaccin est moins neurovirulent et moins neuroinvasif que les souches virulentes de JEV en modèle de souris, et protège contre des infections humaines par le JEV. Cependant, des cas d’encéphalites ont été rapportés après injection de ce vaccin. Il apparait donc que, dans certains cas, la souche vaccinale JEV SA14-14-2 est capable de traverser la BBB et d’infecter les cellules neuronales. Les dernières épidémies à virus ZIKV en Polynésie Française et en Amérique du Sud ont induit une augmentation de cas de malformations congénitales dans les zones touchées. Cela a soulevé de nouvelles questions quant à la capacité d’un Flavivirus à provoquer des malformations congénitales du CNS. Dans cette étude, nous avons identifié les mécanismes cellulaires permettant aux Flavivirus de traverser la BBB et les effets de l’infection par les virus ZIKV et WNV des cellules du CNS sur le développement de celles-ci.Nous avons utilisé deux systèmes in vitro permettant d’étudier le développement du CNS et la neuroinvasion des Flavivirus. Un premier système consiste en l’infection de coupes de cerveaux d’embryon de souris. En utilisant ce système, nous avons montré que le ZIKV a un tropisme préférentiel pour les cellules progénitrices de neurones, alors que le WNV a un tropisme préférentiel pour les neurones. Nous avons également montré que l’infection des progéniteurs neuronaux par le ZIKV induit un arrêt de la mitose cellulaire, alors que l’infection par le WNV n’a aucun effet sur la mitose. L’étude sur l’effet apoptotique de l’infection par les deux virus WNV et ZIKV n’a montré aucune différence entre les deux virus à des temps précoces d’infection.Un deuxième système a été mis au point pour l’étude de la neuroinvasion par les Flavivirus neuropathogènes. Ce système est composé de cellules endothéliales hCMEC/D3 pouvant former des jonctions serrées. Ces cellules ont été cultivées sur filtres d’insert de puits de culture cellulaire Transwell, placés au-dessus de cellules neuronales humaines. A l’aide de ce système, nous avons comparé la capacité à traverser la BBB de plusieurs Flavivirus. / Neuropathogenic Flaviviruses, such as Japanese encephalitis virus (JEV), West Nile virus (WNV), yellow fever virus (YFV) and Zika virus (ZIKV), cause neurological diseases. These diseases are due to viral infection of central nervous system (CNS) cells. The CNS is a privileged organ, isolated from pathogenic agents by a barrier between the blood and the barrier, called the blood-brain barrier (BBB). Neuropathogenic Flaviviruses which can cross this BBB in order to reach their target cells in the CNS, are neuroinvasive. This study aims at understanding the cellular mechanisms by which YFV and JEV Flaviviruses cross the BBB and the effects of viral infection by WNV and ZIKV of the CNS cells during neocortex development.YFV is a hepatrotopic virus, which mostly infects the liver and the kidneys. The two live-attenuated vaccines against YFV, the FNV (for French Neurotropic Virus) vaccine and the 17D vaccine, were obtained empirically by several passages in suckling mouse brain of YFV virulent strains. These vaccines do not cause any disease targeting the liver or the kidneys, but can sometimes cause post-vaccine encephalitis. These encephalitis cases suggest that the vaccine strains have become neurovirulent and neuroinvasive. Due to high risks of post-vaccine encephalitis, the FNV vaccine use was discontinued in the 1980s.JEV is a neurotropic virus, causing acute encephalitis in South East Asia. To date, there is a live-attenuated vaccine against JEV, the JEV SA14-14-2 vaccine, which was obtained empirically by several passages in primary hamster kidney cells. This vaccine is less neurovirulent and less neuroinvasive than JEV virulent strains in mouse model, and it protects against JEV infections. However, some cases of post-vaccine encephalitis were reported. It thus seems that, in some cases, the vaccine strain JEV SA14-14-2 is able to cross the BBB and infect neuronal cells.The recent ZIKV epidemics in French Polynesia and South America were linked to an increase in the number of congenital malformations, rising questions regarding the capacity of a Flavivirus to induce CNS congenital malformations.In this study, we have identified cellular mechanisms involved in Flavivirus neuroinvasion and studied the effect of ZIKV and WNV infection of neuronal cells under development.To study CNS development, we have infected mouse embryos brain slices. We were able to show that ZIKV has a preferential tropism for neuronal progenitors, whereas WNV has a preferential tropism for neuronal cells. We also show that infection of neuronal progenitors by ZIKV impairs the cell life cycle, whereas no effect on the cell life cycle was observed for WNV-infected cells. Studies on apoptosis induction did not show any difference between both viruses at early time points of infection.To study Flavivirus neuroinvasion, we have used an in vitro model of BBB composed of human endothelial hCMEC/D3 cells that can form tight junctions. These cells were cultivated on Transwell inserts and placed above human neuronal cells. Using this system, we show that YFV FNV cross the BBB more efficiently than YFV 17D, suggesting that YFV FNV is more neuroinvasive than YFV 17D. This observation can explain the higher post-vaccine encephalitis risks associated with YFV FNV vaccine compared to YFV 17D vaccine. We also confirmed that JEV SA14-14-2 vaccine strain is less neuroinvasive than JEV RP9.We also examined how JEV crosses the BBB and the endothelial cell response following JEV treatment. We show that both JEV RP9 and SA14-14-2 are able to cross the BBB without infecting its endothelial cells and without disrupting the BBB. Preliminary results suggest that JEV RP9, but not JEV SA14-14-2, crosses the BBB by dynamin-dependant transcytosis. Transcriptomic analysis of endothelial cells treated by either virus show slight, but significant, differences in regulation of genes implicated in several pathways associated with CNS diseases.

Huntington’s disease (HD) is an autosomal dominant, progressive neurodegenerative disorder. Invariably fatal, HD is caused by expansion of the CAG repeat region in exon 1 of the Huntingtin gene which creates a toxic protein with an extended polyglutamine tract 1. Silencing mutant Huntingtin messenger RNA (mRNA) is a promising therapeutic approach 2-6. The ideal silencing strategy would reduce mutant Huntingtin while leaving the wild-type mRNA intact. Unfortunately, targeting the disease causing CAG repeat expansion is difficult and risks targeting other CAG repeat containing genes. We examined an alternative strategy, targeting single nucleotide polymorphisms (SNPs) in the Huntingtin mRNA. The feasibility of this approach hinges on the presence of a few common highly heterozygous SNPs which are amenable to SNP-specific targeting. In a population of HD patients from Europe and the United states, forty-eight percent were heterozygous at a single SNP site; one isoform of this SNP is associated with HD. Seventy-five percent of patients are heterozygous at least one of three frequently heterozygous SNPs. Consequently, only five allele-specific siRNAs are required to treat three-quarters of the patients in the European and U.S. patient populations. We have designed and validated siRNAs targeting these SNPs. We also developed artificial microRNAs (miRNAs) targeting Huntingtin SNPs for delivery using recombinant adeno-associated viruses (rAAVs). Both U6 promoter driven and CMV promoter driven miRNAs can discriminate between matched and mismatched targets in cell culture but the U6 promoter driven miRNAs produce the mature miRNA at levels exceeding those of the vast majority of endogenous miRNAs. The U6 promoter driven miRNAs can produce a number of unwanted processing products, most likely due to a combination of overexpression and unintended export of the pri-miRNA from the nucleus. In contrast, CMV-promoter driven miRNAs produce predominantly a single species at levels comparable to endogenous miRNAs. Injection of recombinant self complementary AAV9 viruses carrying polymerase II driven Huntingtin SNP targeting miRNAs into the striatum results in expression of the mature miRNA sequence in the brain and has no significant effect on endogenous miRNAs. Matched, but not mismatched SNP-targeting miRNAs reduce inclusions in a knock-in mouse model of HD. These studies bring us closer to an allele-specific therapy for Huntington’s disease.

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by a loss of voluntary movement over time, leading to paralysis and death. While 10% of ALS cases are inherited or familial (FALS), the majority of cases (90%) are sporadic (SALS) with unknown etiology. Approximately 20% of FALS cases are genetically linked to a mutation in the anti-oxidizing enzyme, superoxide dismutase (SOD1). SALS and FALS are clinically indistinguishable, suggesting a common pathogenic mechanism exists for both types. Since such a large number of genetic mutations in SOD1 result in FALS (>170), it is reasonable to suspect that non-genetic modifications to SOD1 induce structural perturbations that result in ALS pathology as well. In fact, misfolded SOD1 lacking any genetic mutation was identified in end stage spinal cord tissues of SALS patients using misfolded SOD1-specific antibodies. In addition, this misfolded WT SOD1 found in SALS tissue inhibits axonal transport in vitro, supporting the notion that misfolded WT SOD1 exhibits toxic properties like that of FALS-linked SOD1. Indeed, aberrant post-translational modifications, such as oxidation, cause WT SOD1 to mimic the toxic properties of FALS-linked mutant SOD1. Based on these data, I hypothesize that modified, misfolded forms of WT SOD1 contribute to SALS disease progression in a manner similar to FALS linked mutant SOD1 in FALS. The work presented in this dissertation supports this hypothesis. Specifically, one common misfolded form of SOD1 is defined and exposure of this toxic region is shown to enhance SOD1 toxicity. Preventing exposure, or perhaps stabilization, of this “toxic” region is a potential therapeutic target for a subset of both familial and sporadic ALS patients. Further, the possibility of exploiting this misfolded SOD1 species as a biomarker is explored. For example, an over-oxidized SOD1 species was identified in peripheral blood mononuclear cells (PBMCs) from SALS patients that is reduced in controls. Moreover, 2-dimensional gel electrophoresis revealed a more negatively charged species of SOD1 in PBMCs of healthy controls greatly reduced in SALS patients. This species is hypothesized to be involved in the degradation of SOD1, further implicating both misfolded SOD1 and altered protein homeostasis in ALS pathogenesis.

GM1 gangliosidosis is an autosomal recessive lysosomal storage disease, caused by a deficiency in the enzyme β-galactosidase. The disease affects the CNS, liver, kidney, heart and skeletal system, leading to severe neurodegeneration and death. We propose to treat this disorder using ex vivo hematopoietic stem cell therapy. The effectiveness of this therapy requires the recruitment of transduced donor cells to the CNS. This is only found to occur after mice are conditioned with total body irradiation, due to the increase in CNS cytokine production and blood brain barrier permeability that occurs. As the use of total body irradiation in pediatric patients has been linked to future developmental problems, this myeloablation approach is often avoided in younger patients in favor of a conditioning regimen using the chemotherapy drugs, busulfan and cyclophosphamide. Whether donor cells can enter the CNS when a busulfan and cyclophosphamide conditioning regimen is used has not been determined. In this study we plan to quantify the cytokine and blood-brain barrier permeability increases necessary for donor cells to be recruited to the CNS after total body irradiation. We will then investigate whether busulfan and cyclophosphamide conditioning and/or the chronic neuroinflammation present in GM1 mice can produce similar conditions and facilitate the recruitment of donor hematopoietic stem cells to the CNS. Finally we will assess whether ex vivo hematopoietic stem cell gene therapy is still an effective therapy when busulfan and cyclophosphamide are used for myeloablative conditioning.

Cu, Zn superoxide dismutase (SOD1) is a dimeric, β-sandwich, metalloenzyme responsible for the dismutation of superoxide. Mutations covering nearly 50% of the amino acid sequence of SOD1 have been found to acquire a toxic gain-of-function leading to amyotrophic lateral sclerosis. A hallmark of this disease is the presence of insoluble aggregates containing SOD1 found in the brain and spinal cord. While it is unclear how these aggregates or smaller, precursor oligomeric species may be the source of the toxicity, mutations leading to increased populations of unstable, partially folded species along the folding pathway of SOD1 may be responsible for seeding and propagating aggregation. In an effort to determine the responsible species, we have systematically characterized the stability and folding kinetics of five well studied ALS variants: A4V, L38V, G93A, L106V and S134N. The effect of the amino acid substitutions was determined on a variety of different constructs characterizing the various post-translational maturation steps of SOD1: folding, disulfide bond formation and Zn binding. Zn was found to bind progressively tighter along the folding pathway of SOD1, minimizing populations of monomeric species. In contrast, ALS variants were found to have the greatest perturbation in the equilibrium populations of the folded and unfolded state for the most immature, disulfide-reduced metal-free SOD1. In this species, at physiological temperature, four out of five ALS variants were >50% unfolded. Finally the energetic barriers in the folding and unfolding reaction were studied to investigate the unusually slow folding of SOD1. These results reveal that both unfolding and refolding are dominated by enthalpic barriers which may be explained by the desolvation of the chain and provide insights into the role of sequence in governing the folding pathway and rate.

A seizure is defined as an abnormal electrical discharge from the brain that results in the affected area losing its normal function and reacting uncontrollably. A particular subset of seizures, known as absence seizures, are characterized by brief, paroxysmal losses of consciousness that are associated with bilaterally synchronous 3 Hz spike and wave discharges (SWDs) on electroencephalography (EEG). The optimal way to understand any disease state is to study it within the human. Unfortunately, well controlled experiments in humans are difficult due to small patient populations, treatment medications which alter the seizure, and the ethical problems associated with invasive experimental procedures. Animal models of absence seizures provide a means of avoiding the above difficulties but the model should mimic, as closely as possible, the human condition. The goal of this thesis was to develop an animal model of absence epilepsy that could be used to explore, non-invasively, the underlying mechanisms of absence seizures. Functional magnetic resonance imaging (fMRI) was used to non-invasively monitor brain activity during absence seizures in various animal models. In this dissertation I report the development of a pharmacological rat model of absence seizures for use in fMRI investigations. Imaging was performed after absence seizure induction using γ-butyrolactone (GBL) and it was found that the cortico-thalamic circuitry, critical for the formation of SWDs, showed robust signal changes consistent with electroencephalographic recordings in the same animals. Since a major disadvantage of the GBL rat model is that it produces acute, drug-induced seizures, a genetic rat model with spontaneous absence seizures was subsequently developed for fMRI. EEG-triggered fMRI was used to identify areas of brain activation during spontaneous SWDs in the epileptic WAG/Rij rat strain under awake conditions. Significant signal changes were apparent in several areas of the cortex and several important nuclei of the thalamus. These results draw an anatomical correlation between areas in which there is increased fMRI signal and those where SWDs have been previously recorded using electrophysiologic techniques. One way in which absences differ between humans and both of these rat models is that the SWD frequency in humans is classically 3 Hz while in rats it varies from 7 to 11 Hz. Marmoset monkeys were found to model the human absence seizure condition better than other animals because GBL administration in these non-human primates results in the formation of 3 Hz SWDs. This monkey model was developed for awake functional imaging and changes in signal intensity in the thalamus and sensorimotor cortex correlated with the onset of 3 Hz SWDs. The change in BOLD signal intensity was bilateral but heterogeneous, affecting some brain areas more than others.

Amyotrophic lateral sclerosis (ALS), also known as Lou Gehrig’s disease is a fatal neurodegenerative disease. ALS is typically adult onset and is characterized by rapidly progressive loss of both upper and lower motor neurons that leads to death usually within 3-5 years. About 90% of all the cases are sporadic with no family history while the remaining 10% are familial cases with mutations in several genes including SOD1, FUS/TLS, TDP43 and C9ORF72. FUS/TLS (Fused in Sarcoma/Translocated in Liposarcoma or FUS) is an RNA/DNA binding protein that is involved in multiple cellular functions including DNA damage repair, transcription, mRNA splicing, RNA transport and stress response. More than 40 mutations have now been identified in FUS that account for about 5% of all the familial cases of ALS. However, the exact mechanism by which FUS causes ALS is unknown. While significant progress has been made in understanding the disease mechanism and identifying therapeutic strategies, several questions still remain largely unknown. The work presented here aims at understanding the normal functions of FUS as well as the pathogenic mechanisms by which it leads to disease. Several studies showed the association of mutant-FUS with structures made up of RNA and proteins, called stress granules that form under various stress conditions. However, little is known about the role of endogenous FUS under stress conditions. I have shown that under hyperosmolar conditions, the predominantly nuclear FUS translocates into the cytoplasm and incorporates into stress granules. The response is specific to hyperosmolar stress because FUS remains nuclear under other stress conditions tested, such as oxidative stress, ER stress and heat shock. The response of FUS is rapid, and cells with reduced FUS levels are susceptible to the hyperosmolar stress, indicating a pro-survival role for FUS. In addition to investigating the functions of endogenous wild-type (WT) FUS, the work presented also focuses on identifying the pathogenic mechanism(s) of FUS variants. Using various biochemical techniques, I have shown that ALS-causing FUS variants are misfolded compared to the WT protein. Furthermore, in a squid axoplasm based vesicle motility assay, the FUS variants inhibit fast axonal transport (FAT) in a p38 MAPK dependent manner, indicating a role for the kinase in mutant-FUS mediated disease pathogenesis. Analysis of human ALS patient samples indicates higher levels of total and phospho p38, supporting the notion that aberrant regulation of p38 MAPK is involved in ALS. The results presented in this dissertation 1) support a novel prosurvival role for FUS under hyperosmolar stress conditions and, 2) demonstrate that protein misfolding and aberrant kinase activation contribute to ALS pathogenesis by FUS variants.

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by loss of motor neurons, resulting in progressive muscle weakness, atrophy, paralysis and death within five years of diagnosis. About ten percent of cases are inherited, of which twenty percent are due to mutations in the superoxide dismutase 1 (SOD1) gene. Since the only FDA approved ALS drug prolongs survival by just a few months, new therapies for this disease are needed. Experiments in transgenic ALS mouse models have shown that decreasing levels of mutant SOD1 protein alters and in some cases entirely prevents disease progression. We explored this potential therapeutic approach by using a single stranded AAV9 vector encoding an artificial microRNA against human SOD1 injected bilaterally into the cerebral lateral ventricles of neonatal SOD1G93A mice. This therapy extended median survival from 135 to 206 days (a 50% increase) and delayed hind limb paralysis. Animals remained ambulatory until endpoint, as defined by a sharp drop in body weight. Treated animals had a reduction of mutant human SOD1 mRNA levels in upper and lower motor neurons. As compared to untreated SOD1G93A mice, the AAV9 treated mice also had significant improvements in multiple parameters including the number of motor neurons, diameter of ventral root axons, and degree of neuroinflammation in the spinal cord. These studies clearly show that an AAV9-delivered artificial microRNA is a translatable therapeutic approach for ALS.

N. meningitidis is a common commensal of the human upper respiratory tract and a leading cause of bacterial meningitis and septicemia worldwide. The classical pathway of complement (C) is essential for both naturally acquired and vaccine induced immunity against N. meningitidis. Qualitative and/or quantitative differences in anti-meningococcal antibodies (Abs) in serum is one reason for variations in C-dependent bactericidal Ab activity among individuals. I showed that IgG isolated from select individuals could block killing of group B meningococci by Abs that were otherwise bactericidal. Ligand overlay immunoblots revealed that these blocking IgG Abs were directed against a meningococcal antigen called H.8, Killing of meningococci in reactions containing bactericidal mAbs and human blocking Abs was restored when blocking Ab binding to meningococci was inhibited (or competed for) using either synthetic peptides corresponding to H.8 or a non-blocking mAb against H.8. Further, genetic deletion of H.8 from target organisms abrogated blocking. The Fc region of the blocking IgG was required for blocking because F(ab)2 fragments alone generated by pepsin treatment were ineffective. Blocking required IgG glycosylation; deglycosylation of blocking IgG with peptide:N-glycanase (PNGase) eliminated blocking. C4 deposition mediated by a bactericidal mAb directed against a meningococcal vaccine candidate, called factor H-binding protein (fHbp), was reduced by blocking Ab. Anti-fHbp-mediated C4 deposition was unaffected, however, by deglycosylated blocking IgG. Although preliminary, our data suggests blocking of serum bactericidal activity by human anti-H.8 blocking antibody may require mannan-binding lectin (MBL), which itself is a complement activator. Also, whether MBL recruits a complement inhibitor(s) that facilitates blocking remains to be determined. In conclusion, we have identified H.8 as a meningococcal target for novel blocking antibodies that are commonly found in human serum. Blocking Ab may reduce the efficacy of meningococcal vaccines. We propose that outer membrane vesicle-containing meningococcal vaccines may be more efficacious if purged of subversive immunogens such as H.8.

Spinal muscular atrophy (SMA) is an autosomal recessive neurodegenerative disorder that is the leading genetic cause of infantile death. SMA is caused by homozygous deletion or mutation of the survival of motor neuron 1 gene (SMN1). The SMN2 gene is nearly identical to SMN1, however is alternatively spliced. The close relationship to SMN1 results in SMN2 being a very power genetic modifier of SMA disease severity and a target for therapies. In this study we attempt to characterize novel chemical compounds identified as potential activators of the SMN2 gene. Additionally, we sought to determine the regulatory role individual HDAC proteins use to control expression of full length protein from the SMN2 gene. We used quantitative PCR to determine the effects of novel compounds and shRNA silencing of individual HDACs on the steady state levels of a SMN2-luciferase reporter transcripts. We determined that the compounds identified in multiple reporter high throughput screens increased SMN protein levels via transcriptional activation of the SMN2 gene. Other compounds identified in the same screen functioned post-transcriptionally, possibly stabilizing the SMN protein itself by decreasing degradation. Furthermore, we determined that reduction of individual HDAC proteins was sufficient to increase SMN protein levels in a transgenic reporter system. Knockdown of class I HDAC proteins preferentially activated the reporter by increased promoter transcription. Silencing of class II HDAC proteins maintained transcriptional activity; however silencing of HDAC 5 and 6 also appeared to enhance inclusion of an alternatively spliced exon. This collective work defines a quantitative RNA based protocol to determine mechanism of SMN reporter increase in response to any chosen treatment method. Additionally, this work highlights HDAC proteins 2 and 6 as excellent investigative targets. These data are important to the basic understanding of SMN expression regulation and the refinements of current therapeutic compounds as well as the development of novel SMA therapeutics.