La régulation fine des mécanismes d’élimination des ARN messagers (ARNm) au sein des cellules contribue au contrôle de l’expression génétique ainsi qu’à l’adaptation rapide des niveaux de transcrits en réponse à divers événements cellulaires ou stimuli externes. Elle intervient ainsi dans différents aspects de la physiologie cellulaire : différentiation, prolifération, homéostasie, inflammation ou encore défense anti-parasitaire. Les ARNm eucaryotes matures sont protégés d’une dégradation incontrôlée par une coiffe et une queue poly(A), à chacune de leurs extrémités. Le premier événement amorçant la dégradation des ARNm est le raccourcissement de la queue poly(A) par le complexe CCR4/Not par un processus appelé déadénylation. Ensuite, la coiffe 5’ est éliminée pendant l’étape de « decapping » qui est considérée comme une étape cruciale, irréversible et extrêmement contrôlée, nécessaire à la dégradation rapide du corps du messager par Xrn1. L’étape de “decapping” est effectuée via le recrutement d’un complexe protéique formé de l’enzyme Dcp2 et de son co-activateur essentiel Dcp1. Cependant, ce complexe n’est que peu actif et nécessite de nombreux co-facteurs pour être pleinement efficace. Ces facteurs comprennent l’anneau LSm1-7 qui reconnaît l’extrémité 3’ des ARNm déadénylés et interagit avec Pat1, une protéine plateforme qui recrute l’hélicase Dhh1 et les protéines activatrices du decapping Edc1-2-3. Tous ces facteurs sont organisés au sein d’un réseau d’interaction complexe et dynamique qui, dans certaines conditions, colocalise dans les P-bodies, des foyers cytoplasmiques impliqués dans la dégradation des ARNm et dans la répression de la traduction.Même si de nombreuses études ont révélé l’importance des interactions protéine/protéine dans le processus de decapping, peu d’informations sont disponibles sur les mécanismes moléculaires du recrutement et d’activation de Dcp2 par ses différents co-facteurs. De même, en raison de l’absence de structure de Dcp2 en complexe avec un ARNm coiffé, les détails moléculaires de la reconnaissance et du clivage de la coiffe sont inconnus. Mon projet de thèse a pour but de répondre à ces questions par l’étude fonctionnelle et structurale des acteurs du decapping, en utilisant les protéines de la levure Saccharomyces cerevisiae comme système modèle, puisque la plupart des acteurs du decapping sont conservés au sein des eucaryotes. Dans ce but, j’ai exprimé par génie génétique et isolé la majorité des facteurs impliqués dans le “decapping” et reconstitué plusieurs sous complexes comprenant Dcp2 et ses différents cofacteurs. / MRNA decay is a highly regulated process allowing cells to rapidly adapt their abundance of transcripts to environmental conditions. Eukaryotic mRNAs are protected from uncontrolled decay by a cap structure (m7GpppX) and a poly(A) tail at their 5’ and 3’ ends, respectively. The first event initiating the 5’ to 3’ degradation pathway is the shortening of the poly(A) tail by the CCR4/Not complex through a process known as deadenylation. Then the 5’ cap is degraded during the decapping step, which is considered as a crucial and irreversible step before rapid degradation of RNAs. Decapping is accomplished by the recruitment of a protein complex formed by the Dcp2 catalytic subunit and its activator Dcp1. However, this complex has a low intrinsic decapping activity and requires several accessory factors to be fully efficient. These include the Lsm1-Lsm7 complex that binds to the 3’ end of deadenylated mRNAs and promotes decapping. This complex binds to Pat1, a scaffolding protein recruiting other accessory proteins such as Dhh1 and Edc1-3 proteins (Enhancer of Decapping), which favor decapping. After efficient removal of the cap, Xrn1 (the major cytoplasmic 5’-3’ exonuclease) is recruited and degrades the resulting uncapped RNAs. Interestingly, all these proteins are part of dynamic and multifunctional protein assemblies that, under conditions, localize into cytoplasmic foci known as P-bodies.Although many studies have revealed the importance of these protein/protein interactions, little is known concerning the mechanisms of recruitment and activation of the decapping enzyme by its numerous co-factors. Moreover, in the absence of Dcp2 in complex with a capped RNA, molecular details of cap recognition and cleavage are lacking. My thesis project aims at answering these open questions with the structural and functional studies of the decapping machinery, using yeast Saccharomyces cerevisiae as a model organism, as most of decapping actors are well conserved among eukaryotes. For this purpose, I expressed and purified the majority of the decapping factors and reconstituted several sub-complexes including Dcp2 and its cofactors.

Bancos de tumores foram criados para organizar a coleta, arm azenamento e distribuição de amostras biológicas de pacientes oncológicos, favorecendo seu uso nas pesquis as sobre o cân cer. Amostras ade quadas devem ter RNA, DNA e proteínas de boa qualidade. RNA de boa qualidade deve estar íntegro e puro e DNA deve ter boa c oncentração e pur eza. Basea do em norm as in ternacionais, f oi elaborado e implantado um abrangente sistema de controle de qualidade no banco de tumores do Hospital de Câncer de Barretos, que para fins de estudo foi dividido em banco pré-controle de qu alidade (den ominado b anco pré) e em ban co pós- controle de qualidade (denominado banco pós). Objetivando comparar a qualidade das amostras n os dois bancos, atra vés d a extração d e R NA total e d e DNA (utilizando-se homogeneizador de tecidos e Kits), selecionou-se de forma aleatória 200 a mostras tumorais, distribuídas ig ualitariamente entre mama, co lorreto, estômago, pulmão e tireóide, sendo 100 do banco pré e 100 do banco pós. Para se avaliar a influência do tempo de isquemia fria (tempo entre a excisão do e spécime cirúrgico e o congelamento rápido da amostra armazenada) na qualidade do RNA total de amostras tumorais do banco pós, foram coletadas 200 amostras tumorais, distribuídas igualitariamente entre mama, co lorreto, estômago, pulmão e ti reóide, de 100 doadores diferentes, metade com o tempo de isquemia fria (TIF) de até 30 minutos e a o utra metade do mesmo espécime com TIF de 45 minutos. Extraiu-se RNA total dessas amostras (com maceração manual e Trizol) e comparou-se a sua qualidade, através do núm ero de i ntegridade do RNA (RIN), dentr o dos d ois intervalos de tempo e nas diferentes top ografias. Ao c omparar-se amostras com RIN acima de 7 (consideradas ideais para experimentos de microarray), do banco pré e do b anco pó s, for am enc ontrados 73 (73%) no p rimeiro e 87 (87%) no segundo (p=0,013). Ao comparar-se o intervalo de TIF de até 30 minutos com o de 45 minutos, encontrou-se respectivamente 63 (64,3%) e 3 6 (36%) amostras com RNA total intacto, 11 (11,2%) e 17 (17 %) com RNA tot al parcialmente degradado e 24 (24, 5%) e 47 (47%) com RNA t otal degradado (p<0,001). Amostras tireoidianas e colorretais f oram mais sensíveis ao a umento d o T IF (p=0,006 e p=0,03, respectivamente), e as de estômago e pulmão menos sensíveis (p=0,919 e p=0,384, resp ectivamente). Ao comparar-se a s 200 amostras dos dois ban cos, constatou-se que a grande maioria apresentava boa qualidade, porém o banco pós se destacou ao avaliar-se o número de amostras ideais para estudos de microarray, por provável interferência d o TIF, ainda não controlado no banco pr é. Constatou-se também que algumas amostras do banco pré, armazenadas há mais de ci nco anos em freezer a -80ºC, apresentaram excelente qualidade. O presente estudo também mostrou que o TIF é muito importante para a preservação da qualidade do RNA total, por isso, deve-se sempre respeitar o tempo máximo de 30 minutos. Ainda se observou que a de gradação do RNA é tecido dependente e qu e amostras processadas com homogeneizador de tecidos e extraídas com RNeas y Mini Kit apresentaram melhor q ualidade do RNA, qu e as macer adas manualmente e extraídas com Trizol / Tumor banks were created to or ganize the collection, storage and d istribution of biological samples of cancer pa tients, favoring it\'s use in cancer rese arches. Appropriate samples should have good quality of RNA, DNA and p roteins. RNA of good quality should be intact and pure and DNA should have good concentration and pu rity. Ba sed on international sta ndards, we elabo rated and imp lanted an comprehensive s ystem of qu ality control in the tu mor bank of Ba rretos Cancer Hospital, w hich was divided for st udy purposes i n pre bank quality control (denominated pre bank) and post bank qu ality control (denominated post bank). Aiming to compare the quality of the samples in two banks, through the extraction of total RNA and DNA (b y tissue homogenizer and Kits), we se lected 200 tumor samples in a random way, distributed equally among breast, colorectal, stomach, lung and thyroid, being 100 of the pre-bank and 100 of the post bank. To evaluate the influence o f cold ischem ia time (time b etween t he ex cision o f the su rgical specimen and the fast freezing of the stored sample) in the quality of total of RNA tumor sa mples of th e po st bank , we collected 2 00 t umor s amples, distrib uted equally among breast, colorectal, stom ach, lung and th yroid, fro m 100 different donors, half with the cold ischemia time (CIT) up to 30 minutes and the other ha lf of the sam e specimen with CIT exact ly 45 minutes. We ex tracted total RNA of these samples (with manual maceration and T rizol) and c ompared their qu ality, through the RNA integri ty number (RIN), ins ide tw o intervals of time a nd in different topographies. Comparing samples with RIN above 7 (considered ideals for microarray experiments), of the pre bank and of the post bank, we found 73 (73%) in the first and 87 (87%) in the second (p=0,013). Comparing the interval of CIT up to 30 m inutes with the ex actly 45 minutes, we found respectively 63 (64,3%) and 36 (36%) samples with total RNA intact, 11 (11,2%) and 17 (17%) with total RNA partially degraded and 24 (2 4,5%) and 47 (47%) wit h total RNA de graded (p<0,001). Thyroid and colorectal samples were more sensitive to the increase of CIT (p =0,006 and p=0,03, respectively), a nd s tomach and lun g samples less sensitive (p=0,919 and p=0,384, respectively). C omparing the 200 samples from the two b anks, we v erified that the great ma jority had good qu ality; however the post bank stood out the evaluating number of the id eal samples for m icroarray studies, for probable interference of CIT, still n o controlled in the pre bank. We also verified that some samples of the pre bank, stored more than 5 years in freezer at -80 ºC presented e xcellent qu ality. T he stu dy still sho wed that CIT is ver y important to preserve the quality of total RNA, for that, we sh ould always respect the maximum time of 30 minutes. We still observed that the degradation of RNA is tissue dependent and that samples processed with tissue homogenizer and extracted using RNeasy Mini Kit showed better quality of RNA that macerated manually and extracted with Trizol

L’exosome joue un rôle fondamental dans la dégradation de 3’ en 5’ et la maturation des ARNs chez les eucaryotes. Le "coeur" de l’exosome est composé de 9 sous-unités (EXO9). EXO9 est catalytiquement inactif chez l’homme et la levure, et est associé à deux RNases, Rrp6 et Rrp44, responsables de l’activité exonucléolytique de l’exosome. Mes travaux de thèse démontrent que chez Arabidopsis, le coeur de l’exosome EXO9 possède une activité catalytique intrinsèque. Cette activité est dépendante de la présence de phosphate, produit des nucléosides diphosphates et est réversible. Elle possède de ce fait toutes les caractéristiques d’une activité phosphorolytique. L’activité d’EXO9 est impliquée dans l’élimination de sous-produits de la maturation des ARNr et dans la maturation de l’ARNr 5.8S, deux fonctions typiques de l’exosome. Mes travaux révèlent également que AtRRP44, EXO9 et AtRRP6L2 coopèrent de manière séquentielle pour la maturation de l’ARN 5.8S. Mes travaux de thèse constituent la base de travaux futurs visant à comprendre les rôles de l’activité phosphorolytique de l’exosome chez un organisme eucaryote. / The eukaryotic RNA exosome complex is the main 3’-5’ degradation machinery that plays an essential role in RNA decay, quality control and maturation. The exosome core complex (EXO9) is catalytically inert in yeast and humans, and therefore relies on the catalytic activity of associated RNases, Rrp6 and Rrp44. In this study I demonstrated that EXO9 is catalytically active in Arabidopsis. EXO9’s activity is phosphate-dependent, releases nucleoside diphosphates and is reversible, meeting all criteria of a phosphorolytic activity. Importantly, EXO9’s in vivo substrates include the archetypical exosome substrates, rRNA maturation by-products and 5.8S rRNA precursors. My data show that AtRRP44, EXO9 and AtRRP6L2 sequentially cooperate for the processing of 5.8S rRNA. This work sets a basis for studies aiming at further understanding the biological functions of EXO9’s phosphorolytic activity in a eukaryotic organism.

Bancos de tumores foram criados para organizar a coleta, arm azenamento e distribuição de amostras biológicas de pacientes oncológicos, favorecendo seu uso nas pesquis as sobre o cân cer. Amostras ade quadas devem ter RNA, DNA e proteínas de boa qualidade. RNA de boa qualidade deve estar íntegro e puro e DNA deve ter boa c oncentração e pur eza. Basea do em norm as in ternacionais, f oi elaborado e implantado um abrangente sistema de controle de qualidade no banco de tumores do Hospital de Câncer de Barretos, que para fins de estudo foi dividido em banco pré-controle de qu alidade (den ominado b anco pré) e em ban co pós- controle de qualidade (denominado banco pós). Objetivando comparar a qualidade das amostras n os dois bancos, atra vés d a extração d e R NA total e d e DNA (utilizando-se homogeneizador de tecidos e Kits), selecionou-se de forma aleatória 200 a mostras tumorais, distribuídas ig ualitariamente entre mama, co lorreto, estômago, pulmão e tireóide, sendo 100 do banco pré e 100 do banco pós. Para se avaliar a influência do tempo de isquemia fria (tempo entre a excisão do e spécime cirúrgico e o congelamento rápido da amostra armazenada) na qualidade do RNA total de amostras tumorais do banco pós, foram coletadas 200 amostras tumorais, distribuídas igualitariamente entre mama, co lorreto, estômago, pulmão e ti reóide, de 100 doadores diferentes, metade com o tempo de isquemia fria (TIF) de até 30 minutos e a o utra metade do mesmo espécime com TIF de 45 minutos. Extraiu-se RNA total dessas amostras (com maceração manual e Trizol) e comparou-se a sua qualidade, através do núm ero de i ntegridade do RNA (RIN), dentr o dos d ois intervalos de tempo e nas diferentes top ografias. Ao c omparar-se amostras com RIN acima de 7 (consideradas ideais para experimentos de microarray), do banco pré e do b anco pó s, for am enc ontrados 73 (73%) no p rimeiro e 87 (87%) no segundo (p=0,013). Ao comparar-se o intervalo de TIF de até 30 minutos com o de 45 minutos, encontrou-se respectivamente 63 (64,3%) e 3 6 (36%) amostras com RNA total intacto, 11 (11,2%) e 17 (17 %) com RNA tot al parcialmente degradado e 24 (24, 5%) e 47 (47%) com RNA t otal degradado (p<0,001). Amostras tireoidianas e colorretais f oram mais sensíveis ao a umento d o T IF (p=0,006 e p=0,03, respectivamente), e as de estômago e pulmão menos sensíveis (p=0,919 e p=0,384, resp ectivamente). Ao comparar-se a s 200 amostras dos dois ban cos, constatou-se que a grande maioria apresentava boa qualidade, porém o banco pós se destacou ao avaliar-se o número de amostras ideais para estudos de microarray, por provável interferência d o TIF, ainda não controlado no banco pr é. Constatou-se também que algumas amostras do banco pré, armazenadas há mais de ci nco anos em freezer a -80ºC, apresentaram excelente qualidade. O presente estudo também mostrou que o TIF é muito importante para a preservação da qualidade do RNA total, por isso, deve-se sempre respeitar o tempo máximo de 30 minutos. Ainda se observou que a de gradação do RNA é tecido dependente e qu e amostras processadas com homogeneizador de tecidos e extraídas com RNeas y Mini Kit apresentaram melhor q ualidade do RNA, qu e as macer adas manualmente e extraídas com Trizol / Tumor banks were created to or ganize the collection, storage and d istribution of biological samples of cancer pa tients, favoring it\'s use in cancer rese arches. Appropriate samples should have good quality of RNA, DNA and p roteins. RNA of good quality should be intact and pure and DNA should have good concentration and pu rity. Ba sed on international sta ndards, we elabo rated and imp lanted an comprehensive s ystem of qu ality control in the tu mor bank of Ba rretos Cancer Hospital, w hich was divided for st udy purposes i n pre bank quality control (denominated pre bank) and post bank qu ality control (denominated post bank). Aiming to compare the quality of the samples in two banks, through the extraction of total RNA and DNA (b y tissue homogenizer and Kits), we se lected 200 tumor samples in a random way, distributed equally among breast, colorectal, stomach, lung and thyroid, being 100 of the pre-bank and 100 of the post bank. To evaluate the influence o f cold ischem ia time (time b etween t he ex cision o f the su rgical specimen and the fast freezing of the stored sample) in the quality of total of RNA tumor sa mples of th e po st bank , we collected 2 00 t umor s amples, distrib uted equally among breast, colorectal, stom ach, lung and th yroid, fro m 100 different donors, half with the cold ischemia time (CIT) up to 30 minutes and the other ha lf of the sam e specimen with CIT exact ly 45 minutes. We ex tracted total RNA of these samples (with manual maceration and T rizol) and c ompared their qu ality, through the RNA integri ty number (RIN), ins ide tw o intervals of time a nd in different topographies. Comparing samples with RIN above 7 (considered ideals for microarray experiments), of the pre bank and of the post bank, we found 73 (73%) in the first and 87 (87%) in the second (p=0,013). Comparing the interval of CIT up to 30 m inutes with the ex actly 45 minutes, we found respectively 63 (64,3%) and 36 (36%) samples with total RNA intact, 11 (11,2%) and 17 (17%) with total RNA partially degraded and 24 (2 4,5%) and 47 (47%) wit h total RNA de graded (p<0,001). Thyroid and colorectal samples were more sensitive to the increase of CIT (p =0,006 and p=0,03, respectively), a nd s tomach and lun g samples less sensitive (p=0,919 and p=0,384, respectively). C omparing the 200 samples from the two b anks, we v erified that the great ma jority had good qu ality; however the post bank stood out the evaluating number of the id eal samples for m icroarray studies, for probable interference of CIT, still n o controlled in the pre bank. We also verified that some samples of the pre bank, stored more than 5 years in freezer at -80 ºC presented e xcellent qu ality. T he stu dy still sho wed that CIT is ver y important to preserve the quality of total RNA, for that, we sh ould always respect the maximum time of 30 minutes. We still observed that the degradation of RNA is tissue dependent and that samples processed with tissue homogenizer and extracted using RNeasy Mini Kit showed better quality of RNA that macerated manually and extracted with Trizol

No description available.

La dégradation des ARN est un mécanisme essentiel à la régulation de l’expression des génomes. L’importance de l’uridylation dans les mécanismes de dégradation des ARN commence juste à être appréciée. Cette thèse présente l’étudede l’UTP :RNA Uridylyltransferase 1 (URT1) et de son rôle dans la dégradation des ARN chez Arabidopsis thaliana. L’étude des propriétés catalytiques de URT1 montre que cette uridylyltransférase est intrinsèquement spécifique des UTP et distributive pour les premières uridines ajoutées. URT1 est responsable in vivo de l’uridylation des ARNm après une étape de déadénylation, protégeant leur extrémité 3’ et polarisant la dégradation de 5’ en 3’. URT1 est localisée dans le cytosol au niveau des granules de stress et des processing bodies. Le mécanisme d’adressage de URT1 dans les processing bodies implique une partie de la région N terminale prédite comme intrinsèquement désorganisée, alors que le domainenucléotidyltransférase C terminal semble suffisant pour permettre l’adressage de URT1 au niveau des processing bodies et granules de stress en réponse à un stress thermique. Ces travaux de thèse ont permis de mieux comprendre les mécanismes et les rôles de l’uridylation dans la dégradation des ARNm chez Arabidopsis. Ils ouvrent des perspectives dans l’étude d’autres fonctions de l’uridylation comme l’inhibition de la traduction. / RNA degradation is an essential mechanism for the regulation of genome expression. The importance of uridylation for RNA degradation is just emerging. This thesis presents the study of URT1 (UTP :RNA Uridylyltransferase 1) and its role in RNA degradation in Arabidopsis thaliana. URT1 is an uridylyltransferase intrinsically and strictly specific for UTP and is distributive for the first nucleotides added. URT1 uridylates mRNA in vivo after a deadenylation step. This uridylation protects mRNA’s3’ end from further attacks and polarise degradation in the 5’ to 3’ direction. This protection of 3’ ends by uridylation and its conferred polarity of 5’ to 3’ degradation are also detected in polysomes. Uridylation is therefore likely important in case of cotranslational degradation of mRNAs. A region in URT1’s N terminal region predicted to be intrinsically disorganised is required for addressing URT1 to processing bodies. However, following heat shock, the nucleotidyltransferase domain present in the C terminal region of URT1 is sufficient to address URT1 to both processing bodies and stress granules, This work contributes to a better understanding of the mechanisms and roles of uridylation in RNA degradation in Arabidopsis thaliana. These results also open perspectives for studying other functions of uridylation such as translation inhibition.

The aim of the work was testing of two types of magnetic mikrosheres functionalised with –COOH groups for the isolation of bacterial DNA. Isolation of DNA was carried out from crude lysates of cells prepared from pure culture of Lactobacillus paracassei RL-10 in the presence of binding buffer with 2 M NaCl and 16% PEG 6000. The influence of RNA degradation by enzyme RNase A on the amount of isolated DNA was investigated. It was estimated that RNA degradation affects the amount of DNA isolated. The amount of DNA depended on the type of microparticles. Higher amounts of DNA were isolated using particles with higher content of carboxyl groups. DNA applicable in PCR was isolated using both types of microsheres. In next part of the work, microparticles functionalised with –NH2 groups were used to DNA isolation using electrostatic forces. It was shown that buffer with lower pH is suitable for DNA adsorption onto magnetic microparticles.

Résumé : Chez toutes les cellules, une modulation précise de l’expression des gènes est essentielle afin de réguler adéquatement leur métabolisme et de s’adapter aux changements environnementaux. En effet, c’est l’expression des gènes, plutôt que la séquence d’ADN, qui détermine en grande partie la diversité et la complexité des organismes. Celle-ci dépend principalement des changements dans les niveaux d’ARNs cellulaires résultant de la modification de l’équilibre entre leurs taux relatifs de synthèse et de dégradation. Alors que la régulation transcriptionnelle a été largement étudiée par le passé, des études récentes révèlent que la stabilité de l’ARN joue aussi un rôle important dans le modelage du transcriptome. Toutefois, les mécanismes qui assurent la dégradation précise et sélective des ARNs sont globalement mal compris. Au cours de cette thèse, j’ai utilisé la ribonucléase III de levure Saccharomyces cerevisiae (Rnt1p) comme modèle pour étudier comment des transcrits spécifiques sont ciblés pour la dégradation et évaluer sa contribution à la régulation de l’expression génique. Les résultats indiquent que Rnt1p régule l’expression des gènes en utilisant une spécificité élargie pour des structures tige-boucles d’ARN. En effet, un nouveau motif structurel de Rnt1p permet la discrimination des tige-boucles ayant une séquence spécifique tout en bloquant la liaison à des hélices génériques d’ARN double-brin. D’un autre côté, l’identification des signaux de dégradation de Rnt1p à l’échelle du transcriptome a permis de révéler plus de 384 transcrits clivés par Rnt1p, dont la majorité sont des ARN messagers. En outre, l’impact de la délétion de RNT1 sur l’expression de ces gènes est influencé par les conditions de culture des cellules, ce qui suggère que Rnt1p est un important régulateur conditionnel de l’expression génique. Somme toute, les résultats présentés dans cette thèse démontrent comment des ARNs sont spécifiquement choisis pour la dégradation et soulignent l’importance de la dégradation nucléaire dans la régulation de l’expression génique en réponse à des changements environnementaux. // Abstract : Precise modulation of gene expression is essential for any cell in order to regulate its metabolism and adapt to environmental changes. In fact, it is gene expression, rather than DNA sequence alone, which mostly explains the functional diversity and complexity between the different cell types. As such, gene expression mainly results from changes in the levels of cellular RNAs which are, in turn, dependent on the equilibrium between their relative rates of synthesis and degradation. While transcriptional control has been largely studied in the past, recent publications reveal that changes in RNA stability also play an important role in shaping the transcriptome. Unfortunately though, the mechanisms ensuring precise and selective RNA degradation remains poorly understood. In this thesis, I have used the yeast Saccharomyces cerevisiae ribonuclease III (Rnt1p) as a model to study how specific transcripts are targeted for degradation and evaluate its contribution to the regulation of gene expression. The results indicate that Rnt1p regulates gene expression using a broad specificity for structured RNA stem loops. Indeed, a new structural motif of Rnt1p permits discrimination of hairpins with specific sequence while blocking the binding of the generic RNA duplexes recognized by other members of the RNase III family. This highly specific mode of substrate recognition was found to be easily modulated by a flexible network of protein RNA interactions. On the other hand, transcriptome-wide identification of Rnt1p degradation signals uncovered more than 384 transcripts, including 291 mRNAs. Interestingly, the impact of RNT1 deletion on mRNA expression is modulated by changes in the growth conditions of the cell, indicating that Rnt1p is an important regulator of conditional gene expression. Overall, the results presented in this thesis demonstrate how specific RNAs are selected for degradation and highlight the importance of nuclear RNA decay for fine tuning gene expression in response to changes in growth conditions.

No description available.

No description available.