L’ubiquitin-fold modifier (UFM1) fait partie de la classe 1 de la famille de protéine ubiquitin-like (Ubl). UFM1 et Ub ont très peu d’homologie de séquence, mais partagent des similarités remarquables au niveau de leur structure tertiaire. Tout comme l’Ub et la majorité des autres Ubls, UFM1 se lie de façon covalente à ses substrats par l’intermédiaire d’une cascade enzymatique. Il est de plus en plus fréquemment rapporté que les protéines Ubls sont impliquées dans des maladies humaines. Le gène Ufm1 est surexprimé chez des souris de type MCP développant une ischémie myocardique et dans les îlots de Langerhans de patients atteints du diabète de type 2. UFM1 et ses enzymes spécifiques, UBA5, UFL1 et UFC1, sont conservés chez les métazoaires et les plantes suggérant un rôle important pour les organismes multicellulaires. Le Caenorhabditis elegans est le modèle animal le plus simple utilisé en biologie. Sa morphologie, ses phénotypes visibles et ses lignées cellulaires ont été décrits de façon détaillée. De plus, son cycle de vie court permet de rapidement observer les effets de certains gènes sur la longévité. Ce modèle nous permet de facilement manipuler l’expression du gène Ufm1 et de mieux connaître ses fonctions. En diminuant l’expression du gène ufm-1 chez le C.elegans, par la technique de l’ARN interférence par alimentation, nous n’avons observé aucun problème morphologique grave. Les vers ressemblaient aux vers sauvages et possédaient un nombre de progéniture normal. Cependant, les vers sauvage exposés à l’ARNi d’ufm-1 vivent significativement moins longtemps que les contrôles et ce, de façon indépendante de la voie de signalisation de l’insuline/IGF. Chez le C. elegans la longévité et la résistance au stress cellulaire sont intimement liées. Nous n’avons remarqué aucun effet d’ufm-1 sur le stress thermal, osmotique ou oxydatif, mais il est requis pour la protection contre le stress protéotoxique. Il est également nécessaire au maintien de l’intégrité neuronale au cours du vieillissement des animaux. L’ensemble de nos données nous renseigne sur les fonctions putatives du gène Ufm1. / The ubiquitin-fold modifier (UFM1) is part of the type 1 class of the family of ubiquitin-like protein (Ubl). UFM1 and Ub have very little sequence homology but share remarkable similarities in their tertiary structure. Like Ub and most other UBLS, UFM1 binds covalently to its substrates through an enzymatic cascade. It is frequently reported that UBLs are involved in human diseases. UFM-1 is overexpressed in mice developing a myocardial ischemia and in the islets of patients suffering from type 2 diabetes. UFM1 and its specific enzymes, UBA5, UFL1, and UFC1 are conserved in metazoans and plants suggesting an important role in multicellular organisms. Caenorhabditis elegans is one of the the simplest animal models used in biology. Some features such as morphology, visible phenotypes and cell lineage have completely been described. The short lifecycle of C. elegans makes it easy to observe gene effects on longevity. This model allows us to easily manipulate the expression of the Ufm1 gene and learn more about its putative functions. To study putative functions of Ufm1, we decreased the expression of ufm-1 using RNA interference introduces through feeding. No gross morphological disturbances were observed; worms resembled wild type and had a normal brood size. However, worms exposed to ufm-1 RNAi had a significantly shorter lifespan than the controls. This effect is independent of the insulin/IGF pathway, which is a major axis of longevity genetics. In C. elegans longevity and cellular stress resistance are intimately linked. We have observed no effect of ufm-1 on thermal, osmotic or oxidative stress, but it is required for protection against proteotoxic stress. It is also necessary to maintain neuronal integrity during aging. Together, our results shed light on putative functions of Ufm1 gene.

L’ubiquitin-fold modifier (UFM1) fait partie de la classe 1 de la famille de protéine ubiquitin-like (Ubl). UFM1 et Ub ont très peu d’homologie de séquence, mais partagent des similarités remarquables au niveau de leur structure tertiaire. Tout comme l’Ub et la majorité des autres Ubls, UFM1 se lie de façon covalente à ses substrats par l’intermédiaire d’une cascade enzymatique. Il est de plus en plus fréquemment rapporté que les protéines Ubls sont impliquées dans des maladies humaines. Le gène Ufm1 est surexprimé chez des souris de type MCP développant une ischémie myocardique et dans les îlots de Langerhans de patients atteints du diabète de type 2. UFM1 et ses enzymes spécifiques, UBA5, UFL1 et UFC1, sont conservés chez les métazoaires et les plantes suggérant un rôle important pour les organismes multicellulaires. Le Caenorhabditis elegans est le modèle animal le plus simple utilisé en biologie. Sa morphologie, ses phénotypes visibles et ses lignées cellulaires ont été décrits de façon détaillée. De plus, son cycle de vie court permet de rapidement observer les effets de certains gènes sur la longévité. Ce modèle nous permet de facilement manipuler l’expression du gène Ufm1 et de mieux connaître ses fonctions. En diminuant l’expression du gène ufm-1 chez le C.elegans, par la technique de l’ARN interférence par alimentation, nous n’avons observé aucun problème morphologique grave. Les vers ressemblaient aux vers sauvages et possédaient un nombre de progéniture normal. Cependant, les vers sauvage exposés à l’ARNi d’ufm-1 vivent significativement moins longtemps que les contrôles et ce, de façon indépendante de la voie de signalisation de l’insuline/IGF. Chez le C. elegans la longévité et la résistance au stress cellulaire sont intimement liées. Nous n’avons remarqué aucun effet d’ufm-1 sur le stress thermal, osmotique ou oxydatif, mais il est requis pour la protection contre le stress protéotoxique. Il est également nécessaire au maintien de l’intégrité neuronale au cours du vieillissement des animaux. L’ensemble de nos données nous renseigne sur les fonctions putatives du gène Ufm1. / The ubiquitin-fold modifier (UFM1) is part of the type 1 class of the family of ubiquitin-like protein (Ubl). UFM1 and Ub have very little sequence homology but share remarkable similarities in their tertiary structure. Like Ub and most other UBLS, UFM1 binds covalently to its substrates through an enzymatic cascade. It is frequently reported that UBLs are involved in human diseases. UFM-1 is overexpressed in mice developing a myocardial ischemia and in the islets of patients suffering from type 2 diabetes. UFM1 and its specific enzymes, UBA5, UFL1, and UFC1 are conserved in metazoans and plants suggesting an important role in multicellular organisms. Caenorhabditis elegans is one of the the simplest animal models used in biology. Some features such as morphology, visible phenotypes and cell lineage have completely been described. The short lifecycle of C. elegans makes it easy to observe gene effects on longevity. This model allows us to easily manipulate the expression of the Ufm1 gene and learn more about its putative functions. To study putative functions of Ufm1, we decreased the expression of ufm-1 using RNA interference introduces through feeding. No gross morphological disturbances were observed; worms resembled wild type and had a normal brood size. However, worms exposed to ufm-1 RNAi had a significantly shorter lifespan than the controls. This effect is independent of the insulin/IGF pathway, which is a major axis of longevity genetics. In C. elegans longevity and cellular stress resistance are intimately linked. We have observed no effect of ufm-1 on thermal, osmotic or oxidative stress, but it is required for protection against proteotoxic stress. It is also necessary to maintain neuronal integrity during aging. Together, our results shed light on putative functions of Ufm1 gene.

BACKGROUND: Idiopathic normal pressure hydrocephalus (INPH) is still a syndrome generating more questions than answers. Today, research focuses mainly on two areas: understanding the pathophysiology – especially how the malfunctioning CSF system affects the brain parenchyma – and finding better methods to select patients benefiting from a shunt operation. This thesis targets the aspect of finding better selection methods by investigating the measurability of intracranial pressure via lumbar space, and determining if intraparenchymal measurement of long-term ICP-oscillations (B-waves) could be replaced by short-term measurements of CSF pulse pressure waves via lumbar space. Furthermore, I look into the interaction between the CSF system and the parenchyma itself by investigating how the cortical activity of the brain changes after long-term CSF drainage, and if there is any regress in the suggested ischemia after this intervention. Finally, I examine if the neuronal integrity in the INPH brain is impaired, and if this feature is relevant for the likeliness of improvement after CSF diversion. METHODS: The comparison of intracranial and lumbar pressure was made over a vast pressure interval using our unique CSF infusion technique, and it included ten INPH patients. Pressure was measured via lumbar space and in brain tissue, and the pressures were compared using a general linear model. Short-term lumbar pressure waves were quantified by determining the slope between CSF pulse pressure and mean pressure, defined as the relative pulse pressure coefficient (RPPC). The correlation between RPPC, B-waves and CSF outflow resistance was investigated. In a prospective study, functional MRI was used to assess brain activity before and after long-term CSF drainage of 400 ml of CSF in eleven INPH patients. The functionalities tested included finger movement, memory, and attention. The results were benchmarked against the activity in ten healthy controls to identify the brain areas improving after drainage. The ischemia (Lactate) and neuronal integrity (NAA and Choline) were measured in a similar manner in 16 patients using proton MR spectroscopy, and the improvement of the patients after CSF drainage was based on assessment of their gait. RESULTS: There was excellent agreement between ICP measured in brain tissue and via lumbar space (regression coefficient = 0.98, absolute difference < 1 mm Hg). Adjusting for the separation distance between the measuring devices slightly worsened the agreement, indicating other factors influencing the measured difference as well. RPPC measured via lumbar space significantly correlated to the presence of B-waves, but not to outflow resistance. In the prospective study, controls outperformed patients on clinical tests as well as tasks related to the experiments. Improved behaviour after CSF drainage was found for motor function only, and it was accompanied by increased activation in the supplementary motor area (SMA). No lactate was detected, either before or after CSF drainage. NAA was decreased in INPH patients compared to controls, and the NAA levels were higher in the patients improving after drainage. CONCLUSIONS: ICP can be accurately measured via lumbar space in patients with communicating CSF systems. The close relation between RPPC and B-waves indicates that B-waves are primarily related to intracranial compliance, and that measurement of RPPC via lumbar space could possibly substitute B-wave assessment as selection method for finding suitable patients for shunt surgery. Improvement in motor function after CSF drainage was associated to enhanced activity in SMA, supporting the involvement of the cortico-basal ganglia-thalamo-cortical loop in the pathophysiology of INPH. There was no evidence indicating a widespread low-graded ischemia in INPH; however, there was a neuronal dysfunction in frontal white matter as indicated by the reduced levels of NAA. In addition, the level of neuronal dysfunction was related to the likeliness of improvement after CSF removal, normal levels of NAA predisposing for recovery.