京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第20055号 / 医博第4163号 / 新制||医||1018(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 山田 泰広, 教授 戸井 雅和, 教授 野田 亮 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

L’altération constitutionnelle des gènes BRCA1 et BRCA2 est détectée dans 20 à 30% des formes familiales de cancer du sein. La fréquence de mise en évidence d’une mutation BRCA2 selon des critères généalogiques reste modeste. La définition de caractéristiques tumorales communes aux tumeurs du sein survenant dans un contexte de prédisposition lié à BRCA2 a pour objectif l’identification de caractéristiques propres aux tumeurs BRCA2 permettant de mieux définir les indications de recherche de mutation de ce gène, et l’identification de facteurs impliqués dans la tumorigénèse des cancers du sein liés à BRCA2. L’étude des profils génomiques des tumeurs BRCA2 a caractérisé la délétion récurrente des bras longs des chromosomes 13 et 14. L’analyse supervisée des données d’expression entre les tumeurs BRCA2 et les tumeurs familiales BRCAX a identifié une signature spécifique des tumeurs BRCA2. Les exomes des chromosomes 13 & 14 pour 5 tumeurs informatives et leur ADN constitutionnel ont été séquencés afin d’identifier la ou les cibles des régions délétées. Cette analyse a permis la caractérisation de variants somatiques qui seront à étudier dans une large série de cas BRCA2 et contrôles pour conclure sur leur rôle dans la tumorigénèse liée à BRCA2.La caractérisation de pertes de matériel chromosomique spécifiques aux tumeurs BRCA2, rapportée dans plusieurs études, offre une perspective diagnostique avec le développement d’un test FISH utilisable en pratique clinique pour préciser les indications d’une recherche de mutation du gène BRCA2, mais suggère également la présence de gènes cibles candidats dont l’inactivation est requise lors de la cancérisation mammaire liée à BRCA2. / Germline BRCA1 and BRCA2 mutations account for 20-30% of familial breast cancer. The main indication for BRCA2 screening is a family history, but the mutation detection rate in patients selected this way is low. The identification of characteristics common to BRCA2-associated tumors would improve the criteria used to select patients for BRCA2 screening and could identify factors implicated in BRCA2-mutant breast cancer tumorigenesis. The analysis of BRCA2-mutant breast tumor genomic profiles identified deletions of chromosomes 13q and 14q as a common feature of BRCA2-tumors. Supervised gene expression analysis of BRCA2-mutant breast tumors and familial breast tumors without germline BRCA1 or BRCA2 mutations identified a specific BRCA2 gene signature. Exome sequencing of chromosomes 13q and 14q for 5 BRCA2-mutant tumors, and their associated germline DNA was performed in order to identify the target(s) of the specific genomic deletions in the BRCA2 tumors. This analysis characterized somatic variants that will be screened for in a larger cohort of BRCA2 and control tumors cases to explore their role in BRCA2-mutant breast cancer. Our study identified deletions of chromosomes 13q and 14q as a common feature of tumors with germline BRCA2 mutations, as has been observed in several previous studies. We suggest that FISH analysis for the deletion of these chromosomes would be a rapid and technically feasible first step to select tumors worth screening for germline BRCA2 mutations and we hypothesize that the inactivation of candidate genes located in these deleted regions allows the cell to resume division and progress thus contributing to tumorigenesis in BRCA2-mutant tumors.

La rhinosinusite chronique (RSC) est une maladie inflammatoire des sinus suffisamment prévalente au Canada et pouvant avoir un impact significatif sur la qualité de vie des patients. Depuis plusieurs années, cette maladie complexe constitue un terrain fertile pour la recherche scientifique. Bien qu’elle soit bien comprise au niveau syndromique, plusieurs mystères persistent quant aux mécanismes physiopathologiques sous-jacents. Ces questionnements pourraient potentiellement expliquer les différences cliniques observées, particulièrement dans les réponses thérapeutiques. Considérant les lacunes persistantes dans la compréhension entière de la maladie, le présent mémoire tente de faire le point sur les endotypes épithéliaux ainsi que sur le pathobiome de la RSC. Dans un premier temps, l’analyse transcriptomique de cellules épithéliales cultivées en laboratoire de patients avec RSC a permis d’identifier deux groupes distincts en termes de marqueurs inflammatoires et de signature épithéliale. Ceci laisse supposer que ces mécanismes physiopathologiques seraient grandement impliqués dans la maladie. Dans un deuxième temps, et dans un objectif exploratoire et de faisabilité, le séquençage d’échantillons de RSC a permis de détecter la présence de virus, laissant présager que ceux-ci pourraient avoir un rôle dans la physiopathologie de la maladie et combler les mystères résiduels dans sa compréhension globale. En conclusion, les endotypes déterminés par le profil d’expression génique permettraient une meilleure classification de la RSC et une meilleure personnalisation des thérapies. De plus, la présence tangible de virus dans la RSC est certainement prometteuse et ouvre une grande porte de recherche. Toutefois, d’autres études seront nécessaires pour mieux caractériser leur nature et leur rôle. / Chronic rhinosinusitis (CRS) is a complex inflammatory disease of the sinuses that is prevalent in Canada and that can significantly affect patients’ quality of life. For several years now, this heterogenous disease have been in the scope of many scientific research. While CRS is clinically well understood, much remains unknown concerning the underlying pathophysiological mechanisms. This could potentially explain the notable clinical differences observed between patients, especially in terms of therapeutic responses. In hope of bridging the gaps in understanding CRS, the present thesis aims to discuss epithelial endotypes as well as the feasibility and potential role of viruses in CRS. Firstly, gene expression profiling revealed two distinct clusters of CRS patients: a group with severe non-Type 2 inflammation and epithelial dysfunction, and a group characterised by moderate non-Type 2 inflammation and epithelial dysfunction. This highlights that these two different mechanisms are involved in the development of CRS. Secondly, and with an exploratory intent, next-generation sequencing and transcriptomic assembly of CRS samples have confirmed the presence of viruses, suggesting that viruses are present in CRS and could possibly play a role in the pathogenesis of this disease, potentially addressing the shortcomings in CRS. To summarize, epithelial endotypes identified by gene expression profiling could aid in better classifying CRS to provide a tailored therapeutic option. The presence of viruses in CRS samples unfolds an interesting chapter in this field. However, larger populations will be required to better characterize the nature of these viruses and to assess their intrinsic role in CRS.

Background Molecular data of histologically classified oligodendrogliomas are available offering the possibility to stratify these human brain tumors into clinically relevant molecular subtypes. Methods Gene copy number, mutation, and expression data of 193 histologically classified oligodendrogliomas from The Cancer Genome Atlas (TCGA) were analyzed by well-established computational approaches (unsupervised clustering, statistical testing, network inference). Results We applied hierarchical clustering to tumor gene copy number profiles and revealed three molecular subgroups within histologically classified oligodendrogliomas. We further screened these subgroups for molecular glioma markers (1p/19q co-deletion, IDH mutation, gain of chromosome 7 and loss of chromosome 10) and found that our subgroups largely resemble known molecular glioma subtypes. We excluded glioblastoma-like tumors (7a10d subgroup) and derived a gene expression signature distinguishing histologically classified oligodendrogliomas with concurrent 1p/19q co-deletion and IDH mutation (1p/19q subgroup) from those with predominant IDH mutation alone (IDHme subgroup). Interestingly, many signature genes were part of signaling pathways involved in the regulation of cell proliferation, differentiation, migration, and cell-cell contacts. We further learned a gene regulatory network associated with the gene expression signature revealing novel putative major regulators with functions in cytoskeleton remodeling (e.g. APBB1IP, VAV1, ARPC1B), apoptosis (CCNL2, CREB3L1), and neural development (e.g. MYTIL, SCRT1, MEF2C) potentially contributing to the manifestation of differences between both subgroups. Moreover, we revealed characteristic expression differences of several HOX and SOX transcription factors suggesting the activity of different glioma stemness programs in both subgroups. Conclusions We show that gene copy number profiles alone are sufficient to derive molecular subgroups of histologically classified oligodendrogliomas that are well-embedded into general glioma classification schemes. Moreover, our revealed novel putative major regulators and characteristic stemness signatures indicate that different developmental programs might be active in these subgroups, providing a basis for future studies.

Background Molecular data of histologically classified oligodendrogliomas are available offering the possibility to stratify these human brain tumors into clinically relevant molecular subtypes. Methods Gene copy number, mutation, and expression data of 193 histologically classified oligodendrogliomas from The Cancer Genome Atlas (TCGA) were analyzed by well-established computational approaches (unsupervised clustering, statistical testing, network inference). Results We applied hierarchical clustering to tumor gene copy number profiles and revealed three molecular subgroups within histologically classified oligodendrogliomas. We further screened these subgroups for molecular glioma markers (1p/19q co-deletion, IDH mutation, gain of chromosome 7 and loss of chromosome 10) and found that our subgroups largely resemble known molecular glioma subtypes. We excluded glioblastoma-like tumors (7a10d subgroup) and derived a gene expression signature distinguishing histologically classified oligodendrogliomas with concurrent 1p/19q co-deletion and IDH mutation (1p/19q subgroup) from those with predominant IDH mutation alone (IDHme subgroup). Interestingly, many signature genes were part of signaling pathways involved in the regulation of cell proliferation, differentiation, migration, and cell-cell contacts. We further learned a gene regulatory network associated with the gene expression signature revealing novel putative major regulators with functions in cytoskeleton remodeling (e.g. APBB1IP, VAV1, ARPC1B), apoptosis (CCNL2, CREB3L1), and neural development (e.g. MYTIL, SCRT1, MEF2C) potentially contributing to the manifestation of differences between both subgroups. Moreover, we revealed characteristic expression differences of several HOX and SOX transcription factors suggesting the activity of different glioma stemness programs in both subgroups. Conclusions We show that gene copy number profiles alone are sufficient to derive molecular subgroups of histologically classified oligodendrogliomas that are well-embedded into general glioma classification schemes. Moreover, our revealed novel putative major regulators and characteristic stemness signatures indicate that different developmental programs might be active in these subgroups, providing a basis for future studies.