京都大学 / 0048 / 新制・課程博士 / 博士(生命科学) / 甲第17969号 / 生博第295号 / 新制||生||39(附属図書館) / 30799 / 京都大学大学院生命科学研究科統合生命科学専攻 / (主査)教授 米原 伸, 教授 河内 孝之, 教授 朝長 啓造 / 学位規則第4条第1項該当 / Doctor of Philosophy in Life Sciences / Kyoto University / DFAM

The human Absent, small, or homeotic disc1 (ASH1L) is a member of the Trithorax group (TrxG) proteins that play a role in epigenetic gene activation of developmental HOX genes via H3K36me2 methylation mark. ASH1L contains the evolutionarily conserved SET domain responsible for catalyzing monomethylated and dimethylated lysine formation. The crystal structure of the SET domain of ASH1L revealed a substrate-binding pocket blockage caused by an autoinhibitory loop (AI-loop) that undergoes dynamic changes during catalysis and could be exploited for inhibitor development. Studies have shown that the AI-loop regulates the SET domain, thus the KMTase activity of ASH1L. The SET domain adopts an autoinhibited state where the AI-loop blocks the entry of substrate to the active site, have made it a difficult target for the development of inhibitors. The emerging ASH1L's role in multiple oncogenic processes leading to cancer makes it a viable therapeutic target. Effective targeted inhibition of ASH1L enzymatic activity would be a potential therapeutic approach in cancers driven by high HOX gene expression. We employed the state-of-the-art 1H and 13C-detected solution NMR to better understand the ASH1L regulatory mechanism. We investigated the AI-loop's dynamic structure and conformational mobility of backbone and side chains in the absence and presence of the first- in-class small molecule inhibitors. Numerous backbone amide signals across the AI loop and the catalytic cleft of the SET domain are being broadened, indicating the complex interplay of fast local to slow segmental dynamics across the ASH1L SET domain. The binding of the first-in-class inhibitors perturbs the signals around the AI-loop and SAM binding cleft, validating the inhibitor binding site in the solution. The recently published crystal structures of the MRG domain bound to the ASH1L SET domain revealed disordered conformations of the AI-loop and rearrangement in the SAM binding site compared to the apo ASH1L SET domain. It has been proposed that MRG15 allosterically activates ASH1L by releasing the AI loop. Therefore, we performed extensive studies in an aqueous solution to understand the role of MRG15 in stimulating the catalytic activity of ASH1L. We found that the full-length MRG15 is necessary to induce histone methyltransferase activity of the catalytic SET domain of ASH1L. In contrast, the MRG domain alone cannot enhance the catalytic activity. Furthermore, we found that only the complex of ASH1L SET domain with MRG15 but not with isolated MRG domain can interact with nucleosomes. In summary, I have established the direct link between the structural dynamics of the ASH1L SET domain and its enzymatic activity. Moreover, I have defined the adaptor role of the complete MRG15 protein as the substrate recognition factor for the ASH1L protein without perturbing the AI loop or SAM binding site. The atomic level studies mentioned above, supported by the detailed structure and dynamics studies of the first-in-class inhibitor complex with ASH1L, establish the solid foundations for further drug candidate development, selectively targeting the ASH1L and potentially other H3K36me2 methyltransferases.

Chez les eucaryotes, le génome est organisé en chromatine, une structure nucléoprotéique essentielle pour la régulation de l’expression génique ainsi que pour le maintien de la stabilité du génome. Les ciliés sont d’excellents organismes modèles pour étudier les mécanismes généraux qui maintiennent l’intégrité du génomes eucaryote. Chez Paramecium tetraurelia, la différentiation du génome somatique à partir du génome germinal est caractérisée par des événements massifs et reproductibles d’élimination d’ADN. D’une part, des éléments répétés (transposons,régions minisatellites), de plusieurs kilobases de long, sont imprécisément éliminés.D’autre part, 45000 séquences courtes et uniques, appelées IES, sont précisément éliminées au nucléotide près. Une classe de petits ARN, appelé scnRNAs, est impliquée dans la régulation epigénétique de l’élimination d’ADN, mais comment les scnRNA contrôlent l’élimination d’ADN reste mystérieux. Nous avons testé l’hypothèse selon laquelle une organisation particulière de la chromatine, en particulier des modifications post-traductionelles des histones associées à des formes répressives de la chromatine, est impliquée dans le processus d’élimination d’ADN. Nous avons montré que la triméthylation de l’histone H3 sur la lysine 9 et la lysine 27 (H3K9me3 et H3K27me3)apparaît transitoirement dans le noyau somatique en développement au moment où se produisent les événements d’élimination d’ADN. Nous avons identifié la protéine de type Polycomb, Ezl1, et montré qu’elle est une histone methyltransferase qui présente une dualité de substrat et catalyse à la fois la mise en place de K9me3 et K27me3 sur l’histone H3. Nous avons montré que la déposition de H3K9me3 et H3K27me3 dans le noyau en développement requiert les scnRNAs. Des analyses de séquençage haut débit ont montré que Ezl1 est requise pour l’élimination des longues séquences répétées germinales, suggérant que les scnRNA guident la déposition des marques d’histones au niveau de ces séquences. Au contraire des régions répétées du génome, les IES montrent une sensibilité différente aux scnRNAs et à Ezl1, suggérant que plusieurs voies partiellement chevauchantes sont impliquées dans leur élimination. Notre étude montre que des caractéristiques intrinsèques des séquences d’ADN, telles que leur taille, peut contribuer à la définition des séquences germinales à éliminer. De manière intéressante, nous avons aussi montré que Ezl1 est requise pour la répression transcriptionnelle des éléments transposables. Nous suggérons que les voies H3K9me3et H3K27me3 coopèrent et contribuent à préserver le génome somatique de Paramecium des parasites génomiques. / Eukaryotic genomes are organized into chromatin, a complex nucleoprotein structureessential for the regulation of gene expression and for maintaining genome stability.Ciliates provide excellent model organisms with which to gain better understandinginto the regulation of genome stability in eukaryotes. In the ciliate Parameciumtetraurelia, differentiation of the somatic genome from the germline genome ischaracterized by massive and reproducible programmed DNA elimination events. Longregions of several kilobases in length, containing repeated sequences and transposableelements are imprecisely eliminated, whereas 45,000 short, dispersed, single-copyInternal Eliminated Sequences (IESs) are precisely excised at the nucleotide level. Aspecific class of small RNAs, called scnRNAs, is involved in the epigenetic regulation ofDNA deletion. How scnRNAs may guide DNA elimination in Paramecium remains tobe discovered. Here, we investigated whether chromatin structure, in particular histonepost-translational modifications known to be associated with repressive chromatin,might control DNA elimination. We showed that trimethylated lysine 9 and 27 onhistone H3 (H3K9me3 and H3K27me3) appear in the developing somaticmacronucleus when DNA elimination occurs. We identified the Polycomb-groupprotein, Ezl1, and showed that it is a dual histone methyltransferase that catalyzes bothH3K9me3 and H3K27me3 in vitro and in vivo. Genome-wide analyses show thatscnRNA-mediated H3K9me3 and H3K27me3 deposition is necessary for theelimination of long, repeated germline DNA. Conversely, single copy IESs displaydifferential sensitivity to depletion of scnRNAs and Ezl1, unveiling the existence ofpartially overlapping pathways in programmed DNA elimination. Our study revealsthat cis-acting determinants, such as DNA length, also contribute to the definition ofgermline sequences to delete. We further showed that Ezl1 is required fortranscriptional repression of transposable elements. We suggest that H3K9me3 andH3K27me3 pathways cooperate and contribute to safeguard the Paramecium somaticgenome against intragenomic parasites.

Les chondrosarcomes sont des tumeurs malignes osseuses, considérés comme radio- et chimio-résistants, du fait de leur environnement hypoxique. Dans ce contexte, cette étude vise à mieux comprendre le rôle de l’hypoxie dans la résistance de ces tumeurs à la chimiothérapie (cisplatine) et à la radiothérapie (rayons X) et à identifier de nouvelles stratégies thérapeutiques permettant de sensibiliser les chondrosarcomes aux traitements, par un ciblage épigénétique de la méthylation de la lysine 27 de l’histone H3 (H3K27).Dans un premier temps, nous avons montré que, contrairement à ce qui est communément admis, l’hypoxie n’a pas d’effet sur la sensibilité au cisplatine ou aux rayons X dans certains chondrosarcomes alors qu’il augmente la résistance au cisplatine et la sensibilité aux rayons X uniquement dans une lignée de chondrosarcome. Dans un second temps, nous avons montré que le 3-deazaneplanocine A (DZNep) induit l’apoptose dans ces tumeurs, par un mécanisme indépendant de la méthylation de H3K27 et de sa méthylase EZH2 et semblerait agir par la voie Rhoβ/EGFR. Cependant, il provoque des effets secondaires sur la fertilité masculine. Par ailleurs, son association avec le cisplatine potentialise ses effets toxiques sur les chondrosarcomes. Le GSK-J4, quant à lui ralentit la croissance cellulaire des chondrosarcomes et son association avec le cisplatine augmente cet effet. Cette étude souligne que les chondrosarcomes possèdent des mécanismes de régulation cellulaires différents, d’où l’importance de mener des études sur plusieurs lignées cellulaires afin de mieux prédire la réponse aux traitements. De plus, ces travaux démontrent les propriétés anti-tumorales du DZNep et du GSK-J4 dans le traitement de ces tumeurs. / Chondrosarcomas are bone malignant tumors, considered as radio- and chemo-resistant, due to their hypoxic environment. In this context, this study aimed to better understand the role of hypoxia in the resistance of these tumors to chemotherapy (cisplatin) and radiotherapy (X-rays) and to identify new therapeutic strategies to re-sensitize chondrosarcomas by epigenetic targeting of H3K27 methylation. First, we showed that, contrary to what is commonly accepted, hypoxia has differential effect on cisplatin or X-ray sensitivity in chondrosarcomas, while it increases cisplatin resistance and X-ray sensitivity only in one cell line. Secondly, 3-deazaneplanocin A (DZNep) induces apoptosis in these tumors by a mechanism independent of H3K27 methylation and its methylase EZH2 and seems to act through the Rhoβ / EGFR pathway. However, it causes side effects on male fertility. In addition, its association with cisplatin potentiates its toxic effects on chondrosarcomas. The GSK-J4, on the other hand, decreases cell growth and its association with cisplatin increases this effect.This study highlights that chondrosarcomas use different cellular regulation mechanisms, showing the importance of conducting studies on several cell lines in order to better predict the response to treatments. In addition, these studies demonstrate the anti-tumoral properties of DZNep and GSK-J4 in the treatment of these tumors.

Vertebrate genomes are punctuated by CpG islands regions, which have an elevated frequency of CpG dinucleotides. CpG islands are associated with over 70% of mammalian promoters suggesting they may contribute to the regulation of transcription. However, despite being discovered over 30 years ago, the function of CpG islands is still not understood. Unlike the majority of the genome, CpG islands are resistant to DNA methylation. This provides a binding site for CFP1 which binds specifically to non-methylated DNA via its zinc-finger CXXC (zf-CXXC) domain. CFP1 is a subunit of the SET1 methyltransferase complex, and is thought to direct the activating histone modification H3K4me3 to CpG islands. Interestingly, CFP1 also contains a PHD domain which is proposed to bind the H3K4me3 mark, potentially producing a feedback loop between H3K4me3 and the SET1 complex. Although the structural basis for discrimination of non-methylated CpGs is known, it is not clear how zf-CXXC proteins distinguish CpG islands amongst the irregular nucleosomal landscape which exists within the nucleus. This thesis is focused on the role of CFP1 in the relationship between CpG islands, SET1 and H3K4me3. To address these questions, it was important to mechanistically dissect the contribution of the PHD and zf-CXXC domains. The proposal that the PHD domain of CFP1 binds selectively to H3K4me3 was confirmed by in vitro experiments, however this study demonstrates that the PHD domain is insufficient for stable interactions with chromatin. Using complementary genome-wide and live cell imaging approaches, the zf-CXXC domain shown to be required for PHD-dependent interactions. Genome-wide snapshots of binding interactions, together with spatial and temporal details, expose a surprising contribution of the SET1 complex to the nuclear mobility of CFP1, providing a new perspective on the role of CFP1 in H3K4 methylation.

Différents rôles de l'hélicase Pif1 ont été décrit dont le plus documenté est de décrocher la télomérase des télomères en déroulant les hybrides ARN/ADN formés entre l'ARN de la télomérase et l'ADN télomérique. Plus récemment, une nouvelle voie de signalisation des dommages à l'ADN a été mise en évidence, qui inhibe l'action de la télomérase au niveau d'une cassure de l'ADN via la phosphorylation de l'hélicase Pif1. Cette phosphorylation, dépendante de la kinase ATR (Mec1), inhibe la réparation aberrante de la cassure d'ADN par la télomérase. Nous étudions au sein de l’équipe la protéine RPA (Replication Protein A), affine de l'ADN simple-brin, qui recrute à la fois la protéine de recombinaison homologue Rad52 et la protéine Mec1 impliquée dans la cascade de signalisation des dommages de l'ADN. Lors de l'étude de différentes fonctions de l'hélicase Pif1, j'ai mis en évidence une interaction robuste entre Pif1 et RPA. J'ai identifié un allèle de RFA1, rfa1-D228Y, affectant l'interaction Pif1/RPA et montré, grâce à cet allèle, que cette interaction est impliquée dans le recrutement de Pif1 au niveau d'une cassure double-brins (CDB) induite de l'ADN. Enfin, il a été récemment mis en évidence un nouveau rôle de Pif1 dans la stabilité des G-Quadruplexes durant la réplication du brin avancé. En effet, les cellules pif1 présentent un taux d'instabilité du minisatellite CEB1 inséré sur le brin avancé d'environ 56%, correspondant à des réarrangements de l'ADN de type contractions ou expansions. Lors de l'étude de l'interaction Pif1/RPA, j'ai montré que la mutation rfa1-D228Y entraîne une instabilité du minisatellite CEB1 présent sur le brin avancé, similaire à celle observée avec la délétion pif1∆. Nous suggérons un modèle selon lequel RPA recruterait Pif1 au cours de différents processus cellulaires tels que la réponse des dommages à l'ADN ou la réplication des structures particulières de l'ADN telles que les G-Quadruplexes.En parallèle de cette étude, j’ai étudié le rôle de l'histone méthyltransférase Set1 spécifique de la lysine 4 de l'histone H3 dans la régulation de la taille des télomères. J’ai mis en évidence que le raccourcissement des télomères observé dans un mutant set1 est lié à l'absence de di- et tri-méthylation de H3K4 alors que la perte de monométhylation n'a aucun effet. Cependant, le défaut de la taille des télomères dans les cellules set1∆ n'est pas uniquement lié au défaut de méthylation de H3K4 mais semble impliquer une autre activité de Set1 qu’il reste à déterminer. Etonnamment, nous avons observé que la délétion de SET1 aggrave le raccourcissement des télomères des mutants dont les gènes sont impliqués dans la régulation positive de la taille des télomères et inversement, aggrave le rallongement des télomères de mutants dont les gènes sont impliqués dans la régulation négative des télomères. Nous postulons que l’inactivation de Set1 pourrait à la fois inhiber l’activation précoce des origines de réplication des régions subtélomériques et conduire à un sur-raccourcissement de la taille des télomères, à la fois affecter la synthèse du brin complémentaire dans un contexte où celle-ci est affectée (mutant rif1) et conduire à un sur-allongement des télomères. Une seconde hypothèse propose que Set1 régulerait la transcription deTERRA dans des cellules ayant les télomères déprotégés (mutant rif) entraînant le sur-allongement des télomères. / Different roles of Pif1 helicase have been described, the best documented being to remove telomerase from telomeres by unwinding the RNA/DNA hybrid between telomerase RNA and telomeric DNA. Recently, it was shown that the DNA damage signaling down-regulates telomerase action at a DNA break via Pif1 phosphorylation. Pif1 phosphorylation is dependent of the checkpoint kinase ATR (Mec1) and prevents the aberrant healing of broken DNA ends by telomerase. In our laboratory, we study RPA (Replication Protein A), a single-strand DNA binding protein which recruits the proteins involved in the DNA damage response and checkpoint regulation, such as the homologous recombination protein Rad52 and Mec1 involved in the DNA damage response. I have identified an allele of RFA1, rfa1-D228Y, that affects the Pif1/RPA interaction and showed using this allele that this interaction is implicated in the Pif1 recruitment at an induced double-strand break. Recently, a new role of Pif1 in the stability of G-quadruplex DNA during the leading strand replication has been described. pif1 cells show an instability about 56% of the human minisatellite CEB1 inserted on the leading strand. During my study of the Pif1/RPA interaction, I showed that the rfa1-D228Y mutant induced a similar instability of CEB1 minisatellite on the leading strand. We suggested that RPA would recruit Pif1 for many cellular processes such as DNA damage response or replication of secondary DNA structures such as G-Quadruplexes.In parallel, I have studied the role of the Set1 Histone methyltransferase which catalyse the methylation of the lysine 4 of histone H3, in the regulation of telomere length. I showed that the telomere shortening observed in set1 mutant is due to the loss of di- and tri-methylation of H3K4 while the loss of monomethylation has no effect. However, the short telomeres in set1∆ cells is not only due to the methylation defect shedding light on a new Set1 activity that remains to be fully characterized.. The SET1 deletion aggravates the telomere shortening of mutants which genes are involved in positive regulation of telomere length and conversely, aggravates the lengthening of mutants which genes are involved in negative regulation of telomere length. We postulated that inactivation of Set1 could affect at once activation of early-replication origins and leads to a telomere shortening, and affect synthesis of complementary strand in a context where this one is affected (mutant rif1) and leads to a telomere lengthening. A second hypothesis propose that Set1 would regulate TERRA transcription in cells with deprotected-telomere (rif mutant) leading to the lengthening of telomeres.

Small molecules are important not only for treating human diseases but also for studying disease-related biological processes. This dissertation focuses on the effects of small molecules on pancreatic ductal adenocarcinoma cells. Here, I describe the discovery of two small-molecule tool compounds and their applications for interrogating the biological processes related to two distinct diseases in the human pancreas. First, BRD4770 was identified as a histone methyltransferase inhibitor through a target-based biochemical approach, and was used as a probe to study the function of methyltransferases in cancer cells. Second, BRD7552 was discovered as an inducer of Pdx1 using a cell-based phenotypic screening approach, and was used to induce the expression of Pdx1, a master regulatory transcription factor required for \(\beta\)-cell transdifferentiation. This compound is particularly interesting for the study of type-1 diabetes (T1D). The histone methyltransferase G9a catalyzes methylation of lysine 9 on histone H3, a modification linked to aberrant silencing of tumor-suppressor genes. The second chapter describes the collaborative effort leading to the identification of BRD4770 as a probe to study the function of G9a in human pancreatic cancer cells. BRD4770 induces cellular senescence and inhibits both anchorage-dependent and -independent proliferation in PANC-1 cell line, presumably mediated through ATM-pathway activation. Chapter three describes the study of a natural product gossypol, which significantly enhances the BRD4770 cytotoxicity in p53-mutant cells through autophagic cell death. The up-regulation of BNIP3 might be responsible for the synergistic cell death, suggesting that G9a inhibition may help overcome drug resistance in certain cancer cells. Ectopic overexpression of Pdx1, Ngn3, and MafA can reprogram pancreatic exocrine cells to insulin-producing cells in mice, which sheds light on a new avenue for treating T1D. The fourth chapter focuses on a gene expression-based assay using quantitative real-time PCR technique to screen >60,000 compounds for induction of one or more of these three transcription factors. A novel compound BRD7552 which up-regulated Pdx1 mRNA and protein levels in PANC-1 cells was identified. BRD7552 induces changes of the epigenetic markers within the Pdx1 promoter region consistent with transcriptional activation. Furthermore, BRD7552 partially complements Pdx1 in cell culture, enhancing the expression of insulin induced by the introduction of the three genes in PANC-1 cells. In summary, the central theme of my dissertation is to identify novel bioactive small molecules using different screening approaches, as well as to explore their effects in pancreatic ductal cells. / Chemistry and Chemical Biology

Major Histocompatibility Complex Class-II (MHC-II) molecules are critical regulators of adaptive immunity that present extracellular antigens required to activate CD4+ T cells. MHC-II are regulated at the level of transcription by master regulator, the Class II Transactivator (CIITA), whose association with the MHC-II promoter is necessary to initiate transcription. Recently, much research focused on novel mechanisms of transcriptional regulation of critical genes like MHC-II and CIITA; findings that the macromolecular complex of the 26S-proteasome is involved in transcription have been perhaps the most exciting as they impart novel functions to a well studied system. Proteasome is a multi-subunit complex composed of a 20S-core particle capped by a 19S-regulatory particle. The 19S contains six ATPases which are required for transcription initiation and elongation. We demonstrate that 19S ATPase-S6a inducibly associates with CIITA promoters. Decreased expression of S6a negatively impacts recruitment of the transcription factors STAT-1 and IRF-1 to the CIITA due to significant loss in histone H3 and H4 acetylation. S6a is robustly recruited to CIITA coding regions, where S6a binding coordinates with that of RNA polymerase II. RNAi mediated S6a knockdown significantly diminishes recruitment of Pol II and P-TEF-b components to CIITA coding regions, indicating S6a plays important roles in transcriptional elongation. Our research is focused on the ways in which accessibility to and transcription of DNA is regulated. While cancers are frequently linked to dysregulated gene expression, contribution of epigenetics to cancers remains unknown. To achieve metastatic ability, tumors alter gene expression to escape host immunosurveilance. MHC-II and CIITA expression are significantly downregulated in highly metastatic MDA-MB-435 breast cancer cells. This suppression correlates with elevated levels of the silencing modification H3K27me3 at CIITA and a significant reduction in Pol II recruitment. We observe elevated binding of the histone methyltransferase to CIITApIV and demonstrate this enzyme is a master regulator of CIITA gene expression. EZH2 knockdown results in significant increases in CIITA and MHC-II transcript levels in metastatic cells. In sum, transcriptional regulation by the 19S-proteasome and histone modifying enzymes represents novel mechanisms of control of mammalian gene expression and present novel therapeutic targets for manipulating MHC expression in disease.

No description available.

No description available.