Un tiers de la population européenne est touché par au moins un trouble du cerveau. En effet, la maladie d’Alzheimer, et les gliomes, représentent respectivement le syndrome de démence et les tumeurs cérébrales les plus fréquentes chez l’homme. Plusieurs études épidémiologiques ont montré l’existence d’une corrélation inverse entre le risque de développer une maladie neurodégénérative et un cancer cérébral. Ceci suggère la présence de dénominateurs moléculaires communs entre ces pathologies. Dans les deux cas, un dysfonctionnement mitochondrial est rapporté, représentant une caractéristique partagée par ces deux troubles neurologiques. La protéine kinase mitochondriale PINK1 responsable, lorsqu’elle est mutée, d’une forme précoce et familiale de Parkinson, est particulièrement impliquée dans les processus de maintien de l’homéostasie mitochondriale. Par conséquent, les mécanismes moléculaires régulant PINK1 ainsi que leurs impacts au cours des désordres mitochondriaux répertoriés dans la maladie d’Alzheimer et les tumeurs cérébrales, ont suscité un intérêt central, lors de ma thèse. Au cours de ce travail, nous avons examiné certaines des fonctions mitochondriales de PinK1, associées au maintien de l’homéostasie mitochondriale dans un contexte « Alzheimerisé ». Nous mettons en évidence le rôle de la γ-secretase dans la physiologie mitochondriale en contrôlant la régulation transcriptionnelle de PINK1 par l’AICD, le fragment généré conjointement avec le peptide amyloïde toxique Aβ, à partir du précurseur βAPP. Nous montrons de surcroît, l’initiation de cette régulation par la parkine. / One third of the European populations is affected by a brain disorder. Thus, Alzheimer’s disease and gliomas represent the most frequent human brain dementia syndrome and tumor type, respectively. Several epidemiological studies have shown an inverse relationship between the risk of developing a neurodegenerative disease and a brain tumor, suggesting the existence of common molecular denominators between these pathologies. Interestingly, both pathologies are characterized by a mitochondrial dysfunction. The mitochondrial kinase associated to autosomal recessive Parkinson’s disease, PINK1, is particularly implicated in the control of mitochondrial homeostasis. The main objective of my thesis was to study the molecular mechanisms underlying PINK1 gene regulation and their link with the mitochondrial dysfunction observed in either Alzheimer’s disease or gliomas. Thus, during my thesis we have examined the ability of PINK1 to control mitochondria homeostasis in an Alzheimer’s pathological context. We demonstrate that AICD, a cleavage product of the trans-membrane protein βAPP by γ-secretase, impacts mitochondrial physiology via its ability of positively controlling PINK1 transcription. In addition, we show that the signaling cascade linking γ-secretase and PINK1 is initiated by parkin transcriptional regulation of presenilins, the main component of γ-secretase catalytic complex. Finally, we also establish that the tumor suppressor p53 can negatively regulate PINK1 transcription in vitro and in vivo suggesting that the misregulated autophagic response associated to brain tumors development may be caused by defective p53-PINK1 interplay.

<p>Alzheimer Disease (AD) is the most common neurodegenerative disorder in the world. One of the neuropathological hallmarks of AD is the senile plaques in the brain. The plaques are mainly composed of the amyloid β (Aβ) peptide. Aβ is generated from the amyloid precursor protein, APP, when it is first cleaved by the β-secretase and subsequently the γ-secretase complex. The γ-secretase complex cleaves at different sites, called γ and ε, where the γ-cleavage site generates Aβ peptides of different lengths and ε-cleavage generates the APP intracellular domain (AICD). The two major forms of Aβ is 40 and 42 amino acids long peptides, where the latter is more prone to aggregate and is the main component in senile plaques. The γ-secretase complex is composed of four proteins; Pen-2, Aph-1, nicastrin and presenilin (PS). The PS protein harbours the catalytic site of the complex, where two aspartate residues in position 257 and 385 (Presenilin 1 numbering) are situated. Most Familial AD (FAD) mutations in the PS gene cause a change in the γ-cleavage site, leading to a shift from producing Aβ40 to the longer more toxic variant Aβ42. Frequently, this often leads to impairments of the AICD production. Another substrate for the γ-secretase complex is Notch. It is important to maintain the Notch signaling since an intracellular domain (NICD) is formed after cleavage by the γ-secretase complex in the membrane (S3-site) and this domain is involved in transcription of genes important for cell fate decisions.</p><p>It has been reported that certain APP luminal juxtamembrane mutations could drastically alter Aβ secretion, however their effect on AICD production remains unknown. In this study we want to analyse wether the juxtamembrane region is important for the AICD production. To gain more insight into the luminal juxtamembrane function for γ-secretase-dependent proteolysis, we have made a juxtamembrane chimeric construct. A four-residue sequence preceding the transmembrane domain (TMD) of APP (GSNK), was replaced by its topological counterpart from the human Notch1 receptor (PPAQ). The resulting chimeric vector C99GVP-PPAQ and the wildtype counterpart were expressed in cells lacking PS1 and PS2 (BD8) together with PS1wt. We observed that the chimeric construct did not alter production of AICD when using a cell based luciferase reporter gene assay monitoring AICD production. We also introduced a PS1 variant lacking a big portion of the large hydrophilic loop, PS1∆exon10, since our group has previously observed that this region affect Aβ production¹⁴³. We found that the absence of the large hydrophilic loop in PS1 gave a 2-fold decrease in AICD-GVP formation from C99GVPwt compared to PS1wt.  The activity of PS1wt and PS1Δexon10 using C99GVP-PPAQ as a substrate gave similar result as the C99GVPwt substrate, i.e. a 2-fold decrease in AICD-GVP formation when comparing PS1Δexon10 with PS1wt. From this data we therefore suggest that the four residues in the juxtramembrane domain (JMD) (GSNK) is not altering ε-cleavage of APP when changed to Notch1 counterpart, PPAQ. Furthermore, we also show that the 2-fold decrease in AICD-production by the PS1Δexon10 molecule is not changed between the two substrates C99GVPwt and C99GVP-PPAQ. This indicates that the luminal region of APP is not directly involved in the ε-site processing. If the luminal region is affecting processing in the γ-cleavage sites, remains however to be investigated.</p>

Alzheimer Disease (AD) is the most common neurodegenerative disorder in the world. One of the neuropathological hallmarks of AD is the senile plaques in the brain. The plaques are mainly composed of the amyloid β (Aβ) peptide. Aβ is generated from the amyloid precursor protein, APP, when it is first cleaved by the β-secretase and subsequently the γ-secretase complex. The γ-secretase complex cleaves at different sites, called γ and ε, where the γ-cleavage site generates Aβ peptides of different lengths and ε-cleavage generates the APP intracellular domain (AICD). The two major forms of Aβ is 40 and 42 amino acids long peptides, where the latter is more prone to aggregate and is the main component in senile plaques. The γ-secretase complex is composed of four proteins; Pen-2, Aph-1, nicastrin and presenilin (PS). The PS protein harbours the catalytic site of the complex, where two aspartate residues in position 257 and 385 (Presenilin 1 numbering) are situated. Most Familial AD (FAD) mutations in the PS gene cause a change in the γ-cleavage site, leading to a shift from producing Aβ40 to the longer more toxic variant Aβ42. Frequently, this often leads to impairments of the AICD production. Another substrate for the γ-secretase complex is Notch. It is important to maintain the Notch signaling since an intracellular domain (NICD) is formed after cleavage by the γ-secretase complex in the membrane (S3-site) and this domain is involved in transcription of genes important for cell fate decisions. It has been reported that certain APP luminal juxtamembrane mutations could drastically alter Aβ secretion, however their effect on AICD production remains unknown. In this study we want to analyse wether the juxtamembrane region is important for the AICD production. To gain more insight into the luminal juxtamembrane function for γ-secretase-dependent proteolysis, we have made a juxtamembrane chimeric construct. A four-residue sequence preceding the transmembrane domain (TMD) of APP (GSNK), was replaced by its topological counterpart from the human Notch1 receptor (PPAQ). The resulting chimeric vector C99GVP-PPAQ and the wildtype counterpart were expressed in cells lacking PS1 and PS2 (BD8) together with PS1wt. We observed that the chimeric construct did not alter production of AICD when using a cell based luciferase reporter gene assay monitoring AICD production. We also introduced a PS1 variant lacking a big portion of the large hydrophilic loop, PS1∆exon10, since our group has previously observed that this region affect Aβ production143. We found that the absence of the large hydrophilic loop in PS1 gave a 2-fold decrease in AICD-GVP formation from C99GVPwt compared to PS1wt.  The activity of PS1wt and PS1Δexon10 using C99GVP-PPAQ as a substrate gave similar result as the C99GVPwt substrate, i.e. a 2-fold decrease in AICD-GVP formation when comparing PS1Δexon10 with PS1wt. From this data we therefore suggest that the four residues in the juxtramembrane domain (JMD) (GSNK) is not altering ε-cleavage of APP when changed to Notch1 counterpart, PPAQ. Furthermore, we also show that the 2-fold decrease in AICD-production by the PS1Δexon10 molecule is not changed between the two substrates C99GVPwt and C99GVP-PPAQ. This indicates that the luminal region of APP is not directly involved in the ε-site processing. If the luminal region is affecting processing in the γ-cleavage sites, remains however to be investigated.

Previously, we showed that ADAM10 is necessary for HIV-1 replication in primary human macrophages and immortalized cell lines. Silencing ADAM10 expression interrupted the HIV-1 life cycle prior to nuclear translocation of viral cDNA. Furthermore, our data indicated that HIV-1 replication depends on the expression of ADAM15 and γ-secretase, which proteolytically processes ADAM10. Silencing ADAM15 or γ-secretase expression inhibits HIV-1 replication between reverse transcription and nuclear entry. Here, we show that ADAM10 expression also supports replication in CD4+ T lymphocytes. The intracellular domain (ICD) of ADAM10 associates with the HIV-1 pre-integration complex (PIC) in the cytoplasm and immunoprecipitates and co-localizes with HIV-1 integrase, a key component of PIC. Taken together, our data support a model whereby ADAM15/γ-secretase processing of ADAM10 releases the ICD, which then incorporates into HIV-1 PIC to facilitate nuclear trafficking. Thus, these studies suggest ADAM10 as a novel therapeutic target for inhibiting HIV-1 prior to nuclear entry.

All over the world, societies are facing rapidly aging populations combined with a growing number of patients suffering from Alzheimer’s disease (AD). One focus in pharmaceutical research to address this issue is on the reduction of the longer amyloid-β (Aβ) fragments in the brain by modulation of γ-secretase, a membrane-bound protease. R-Flurbiprofen (tarenflurbil) was studied in this regard but failed to show significant improvement in AD patients in a phase 3 clinical trial. This was mainly attributed to its low ability to cross the blood–brain barrier (BBB). Here, we present the synthesis and in vitro evaluation of a racemic meta-carborane analogue of flurbiprofen. By introducing the carborane moiety, the hydrophobicity could be shifted into a more favourable range for the penetration of the blood–brain barrier, evident by a logD7.4 value of 2.0. Furthermore, our analogue retained γ-secretase modulator activity in comparison to racemic flurbiprofen in a cell-based assay. These findings demonstrate the potential of carboranes as phenyl mimetics also in AD research.

アルツハイマー病の原因物質とされるAβの前駆体C99にはGXXXGモチーフ(以下モチーフ)が存在し、二量体形成に関与してAβ産生に影響を及ぼすとされる。そこでアミノ酸置換によりモチーフのないC99を調製し、その二量体形成とAβ産生への影響を検討した。その結果、モチーフは二量体形成には必要でないこと、Aβ産生を担う酵素の切断を制御する可能性が示された。以上より、モチーフが元来言われていた二量体形成よりも、Aβ産生に強く影響を及ぼすことを示唆している。 / Substitutions in GXXXG motifs did not significantly alter C99 dimerization, decreased the production of long Aβ species and increased short Aβ species, and decreased the ratios of precursor/product. In conclusion, my data indicate that GXXXG motifs of C99 are not crucial for the formation of C99 dimers. My findings might be useful for a development of new ways of therapeutic interventions. Because the amino acid substitutions in GXXXG motifs reduced the levels of amyloidogenic Aβ production, drugs targeted to GXXXG motifs would reduce the production of substances of great interest in the field of Alzheimer neuropathology, thereby inhibiting the formation of amyloid plaques. / 博士(理学) / Doctor of Philosophy in Science / 同志社大学 / Doshisha University

La maladie d’Alzheimer (MA) est une pathologie neurodégénérative communément caractérisée par une perte progressive de la mémoire. L’étiologie de la MA demeure incertaine à ce jour ce qui complique l’élaboration de stratégies thérapeutiques permettant de l’éradiquer. L’accumulation des échecs thérapeutiques pourrait en partie s’expliquer par le fait que l’hypothèse amyloïde, qui met en avant l’implication prépondérante du peptide bêta-amyloïde (Aβ) dans la physiopathologie de la MA, serait incomplète. En utilisant un modèle murin transgénique de la MA, la souris TgCRND8, j’ai pu compléter l’hypothèse amyloïde en proposant l’implication, en plus de l’Aβ, du fragment carboxy-terminal bêta (β-CTF). Ces deux métabolites amyloïdogéniques de l’APP seraient responsables de l’altération de formes différentes de mémoire. Le dosage de ces métabolites dans l’hippocampe, suite au traitement chronique des souris avec un inhibiteur de β ou de γ-secrétase, a mis en évidence que le β-CTF serait responsable de l’atteinte de la mémoire impliquée dans la détection du remplacement d’un objet, alors que l’Aβ perturberait la mémoire permettant la détection du déplacement d’un objet. Ces travaux suggèrent qu’il serait judicieux de développer de nouvelles stratégies thérapeutiques qui diminuent à la fois les niveaux cérébraux des deux fragments amyloïdogéniques, le β-CTF et l’Aβ. / Alzheimer’s disease (AD) is a neurodegenerative pathology commonly characterized by a progressive memory loss. To these days, AD’s etiology has remained unclear which complicates the development of therapeutic strategies enabling to eradicate the pathology. The accumulation of therapeutic failures could partly be explained by the fact that the amyloid hypothesis, which highlights the leading involvement of the amyloid beta peptide (Aβ) in the physiopathology of AD, could be incomplete. Using a transgenic mouse model of AD, the TgCRND8 mice strain, I expanded the amyloid hypothesis, suggesting the involvement of the beta carboxy-terminal fragment (β-CTF), in addition to Aβ. These two amyloidogenic metabolites could be responsible for the alteration of different forms of memory. The dosage of these metabolites, after mice chronic treatment with either a β- or a γ-secretase inhibitor, highlighted the fact that β-CTF could be responsible for the deterioration of the memory involved in the detection of the replacement of an object. As for Aβ, it could disrupt the memory allowing the detection of the displacement of an object. This work suggests that it would be judicious to develop therapeutic strategies reducing brain levels of both amyloid fragments, β-CTF and Aβ.