Conselho Nacional de Desenvolvimento Científico e Tecnológico / A transsialidase é uma glicoproteína de membrana pertencente a uma família de genes de cópia múltipla, envolvida no processo de invasão celular do Trypanosoma cruzi no hospedeiro vertebrado. Esta dissertação foi concebida com um amplo componente analítico que dependia de dados publicamente disponíveis, ou seja, as sequências oriundas do projeto genoma de T. cruzi e cDNAs de trans-sialidase depositadas no Genbank-dbEST. Este componente analítico necessitou ser complementado e ampliado com a obtenção experimental de novas sequências, a partir da metodologia baseada na transcrição reversa acoplada a PCR. Os fragmentos obtidos de cepas de T. cruzi Dm28c (T. cruzi I), Y (T. cruzi II), CL-Brener (T. cruzi II, cepa híbrida), INPA4167 (zimodema III), 3663 (zimodema III) e Colombiana (zimodema III) foram clonados, sequenciados e analisados composicionalmente. Essas sequências foram editadas e alinhadas usando-se o software CLUSTAL X. Em uma seção específica do Genbank, denominada dbEST, buscamos os cDNAs homólogos a trans-sialidase. Esta busca por similaridade foi realizada individualmente com os números de acesso referentes às seqüências supracitadas contra o dbEST utilizando o BLAST a fim de obtermos informações funcionais e evolutivas. Em seguida, desenvolvemos metodologias experimentais que nos permitiu avaliar segmentos da 5 UTR, tais como os sítios de trans-splicing adicionais ou múltiplos em TS e seus respectivos sinais (região rica em polipirimidina), variação composicional e tamanho da região das sequências entre diferentes linhagens de T. cruzi. O resultado dessa averiguação também nos mostrou a quantidade de cDNAs relacionados com a transsialidase no dbEST bem como a relação desses cDNAs com o mini-exon. As cepas do zimodema III apresentaram tamanho médio dos fragmentos de 312 bases, enquanto T. cruzi I e T. cruzi II apresentaram, respectivamente 209 e 218. Trans splicing adicional ou duplicações gênicas com mutações no sítio primário de trans splicing não parece ser um fenômeno exclusivo de algum grupo populacional, embora seja mais evidente em T. cruzi zimodema III.

Conselho Nacional de Desenvolvimento Científico e Tecnológico / A transsialidase é uma glicoproteína de membrana pertencente a uma família de genes de cópia múltipla, envolvida no processo de invasão celular do Trypanosoma cruzi no hospedeiro vertebrado. Esta dissertação foi concebida com um amplo componente analítico que dependia de dados publicamente disponíveis, ou seja, as sequências oriundas do projeto genoma de T. cruzi e cDNAs de trans-sialidase depositadas no Genbank-dbEST. Este componente analítico necessitou ser complementado e ampliado com a obtenção experimental de novas sequências, a partir da metodologia baseada na transcrição reversa acoplada a PCR. Os fragmentos obtidos de cepas de T. cruzi Dm28c (T. cruzi I), Y (T. cruzi II), CL-Brener (T. cruzi II, cepa híbrida), INPA4167 (zimodema III), 3663 (zimodema III) e Colombiana (zimodema III) foram clonados, sequenciados e analisados composicionalmente. Essas sequências foram editadas e alinhadas usando-se o software CLUSTAL X. Em uma seção específica do Genbank, denominada dbEST, buscamos os cDNAs homólogos a trans-sialidase. Esta busca por similaridade foi realizada individualmente com os números de acesso referentes às seqüências supracitadas contra o dbEST utilizando o BLAST a fim de obtermos informações funcionais e evolutivas. Em seguida, desenvolvemos metodologias experimentais que nos permitiu avaliar segmentos da 5 UTR, tais como os sítios de trans-splicing adicionais ou múltiplos em TS e seus respectivos sinais (região rica em polipirimidina), variação composicional e tamanho da região das sequências entre diferentes linhagens de T. cruzi. O resultado dessa averiguação também nos mostrou a quantidade de cDNAs relacionados com a transsialidase no dbEST bem como a relação desses cDNAs com o mini-exon. As cepas do zimodema III apresentaram tamanho médio dos fragmentos de 312 bases, enquanto T. cruzi I e T. cruzi II apresentaram, respectivamente 209 e 218. Trans splicing adicional ou duplicações gênicas com mutações no sítio primário de trans splicing não parece ser um fenômeno exclusivo de algum grupo populacional, embora seja mais evidente em T. cruzi zimodema III.

The 5’ untranslated region (UTR) of coronaviral genomes contains cis-acting sequences necessary for replication, transcription and translation. A consensus secondary structural model of the 5' 140 nucleotides of the 5' UTRs of nine coronaviruses (CoVs) derived from all three major CoV groups is presented and characterized by three major stem loops, SL1, SL2 and SL4. SL2 is conserved in all CoVs, typically containing a pentaloop (C47-U48-U49-G50-U51 in MHV) stacked on a 5-bp stem, with some sequences containing an additional U 3' to U51. NMR structural studies of SL2 hairpin reveal that SL2 adopts a U-turn-like conformation. Parallel molecular genetic experiments reveal that SL2 plays an essential role in sgRNA synthesis as does SL1. We observe strong genetic selection against viruses that contain a deletion of A35, an extrahelical nucleotide that destabilizes SL1, in favor of genomes that contain a diverse panel of destabilizing second-site mutations, due to introduction of a collection of non-canonical base pairs near the deleted A35. Viruses containing destabilizing SL1-∆A35 mutations also contain one of two specific single nucleotide mutations in the 3' UTR. Thermal denaturation and imino proton solvent exchange experiments reveal that the lower half of SL1 is unstable and that second-site SL1-∆A35 substitutions recover one or more features of the wild-type SL1. We propose a "dynamic SL1" model that supports viral replication; these characteristics of SL1 appear to be conserved in other coronaviral genomes. The coronaviral nucleocapsid (N) protein contains two or more RNA binding domains. We investigated the RNA-binding properties of the N-terminal (NTD) and Cterminal (CTD) domain of MHV N. Our results reveal that the NTD specifically interacts with the TRS-L3 sequence. The role of conserved residues (Y127, Y129 and R110) for this specific interaction were systematically investigated. In contrast to the NTD, the MHV CTD is homodimeric in solution and binds single-strand RNA nonspecifically in a binding mode of the noncooperative large ligand lattice model. The CTD dimer binds with a site size, n=4 nucleotide and the appending of the NTD enhances the single-strand nucleic acid binding affinity.

The putative attachment protein G of pneumonia virus of mice (PVM), a member of the Pneumoviruses, is an important virulence factor with so far ambiguous function in a virus-cell as well as in virus-host context. The sequence of the corresponding G gene is characterized by significant heterogeneity between and even within strains, affecting the gene and possibly the protein structure. This accounts in particular for the PVM strain J3666 for which two differing G gene organizations have been described: a polymorphism in nucleotide 65 of the G gene results in the presence of an upstream open reading frame (uORF) that precedes the main ORF in frame (GJ366665A) or extension of the major G ORF for 18 codons (GJ366665U). Therefore, this study was designed to analyse the impact of the sequence variations in the respective G genes of PVM strains J3666 and the reference strain 15 on protein expression, replication and virulence. First, the controversy regarding the consensus sequence of PVM J3666 was resolved. The analysis of 45 distinct cloned fragments showed that the strain separated into two distinct virus populations defined by the sequence and structure of the G gene. This division was further supported by nucleotide polymorphisms in the neighbouring M and SH genes. Sequential passage of this mixed strain in the cell line standardly used for propagation of virus stocks resulted in selection for the GJ366665A-containing population in one of two experiments pointing towards a moderate replicative advantage. The replacement of the G gene of the recombinant PVM 15 with GJ366665A or GJ366665U, respectively, using a reverse genetic approach indicated that the presence of uORF within the GJ366665A significantly reduced the expression of the main G ORF on translational level while the potential extension of the ORF in GJ366665U increased G protein expression. In comparison, the effect of the G gene-structure on virus replication was inconsistent and dependent on cell line and type. While the presence of uORF correlated with a replication advantage in the standardly used BHK-21 cells and primary murine embryonic fibroblasts, replication in the murine macrophage cell line RAW 264.7 did not. In comparison, the GJ366665U variant was not associated with any effect on replication in cultured cells at all. Nonetheless, in-vivo analysis of the recombinant viruses associated the GJ366665U gene variant, and hence an increased G expression, with higher virulence whereas the GJ366665A gene, and therefore an impaired G expression, conferred an attenuated phenotype to the virus. To extend the study to other G gene organizations, a recombinant PVM expressing a G protein without the cytoplasmic domain and for comparison a G-deletion mutant, both known to be attenuated in vivo, were studied. Not noticed before, this structure of the G gene was associated with a 75% reduction in G protein expression and a significant attenuation of replication in macrophage-like cells. This attenuation was even more prominent for the virus lacking G. Taking into consideration the higher reduction in G protein levels compared to the GJ366665A variant indicates that a threshold amount of G is required for efficient replication in these cells. In conclusion, the results gathered indicated that the expression levels of the G protein were modulated by the sequence of the 5’ untranslated region of the gene. At the same time the G protein levels modulated the virulence of PVM. / Das mutmaßliche „attachment“ Protein G des Pneumonievirus der Maus (PVM), einem Mitglied des Genus Pneumovirus, ist ein bedeutender Virulenzfaktor, mit allerdings noch nicht vollständig verstandener Funktion. Dabei zeichnet sich die Sequenz des G-Gens durch Nukleotid-Polymorphismen und damit verbundenen Variationen in der Genorganisation und möglicherweise der Proteinstruktur sowohl zwischen als auch innerhalb von PVM-Stämmen aus. Insbesondere für den PVM-Stamm J3666 wurden zwei verschiedene Organisationen des G-Gens beschrieben: ein Polymorphismus des Nukleotids 65 des G-Genes erzeugt einen neuen „upstream Open reading frame“ (uORF), der dem eigentlichen G-ORF vorausgeht (GJ366665A), oder führt zu einer Verlängerung des eigentlichen G-ORF von G um 18 Kodons (GJ366665U). Ziel dieser Studie war es deshalb, die Auswirkung dieser Sequenzvariabilitäten der für PVM J3666 beschriebenen G-Gene im Vergleich zu dem des Referenzstamms PVM 15 bezüglich Proteinexpression, der Virusreplikation und der Virulenz zu untersuchen. Als erstes wurden die beschriebenen Sequenzunterschiede bezüglich des PVM-Stamms J3666 untersucht. Die Analyse von 45 verschiedenen klonierten Fragmenten von PVM J3666 zeigte, dass es sich bei diesem Stamm eigentlich um zwei separate Viruspopulationen handelt, die sich durch die Sequenz und Struktur des G-Genes definieren lassen. Diese Unterscheidung wird durch weitere Nukleotid-Polymorphismen in den benachbarten Genen, M und SH, gestärkt. Sequenzielle Passagierung dieses gemischten Stammes in der standardmäßig zur Virusanzucht verwendeten BHK-21-Zelllinie resultierte in einem von zwei Experimenten in der Selektion der GJ366665A-Population, das ein Hinweis auf einen moderaten Replikationsvorteil darstellt. Der Austausch des G-Gens des Referenzstamms PVM 15 durch GJ366665A oder GJ366665U mithilfe der Reversen Genetik, zeigte, dass der uORF innerhalb von GJ366665A zu einer deutlich reduzierten Expression des eigentlichen G-ORF führt. Andererseits führte die potenzielle Verlängerung des ORF in GJ366665U zu einer im gleichen Maße erhöhten Expression des G-Proteins. Dagegen war der Einfluss der G-Genorganisation auf die Virusvermehrung in Zellkultur in Abhängigkeit von Zelllinie und Zelltyp inkonsistent. Während ein uORF mit einem Replikationsvorteil in BHK-21-Zellen und primären murinen embryonen Fibroblasten korrelierte, war dies in der murinen Makrophagen-Zelllinie RAW 264.7 nicht zu beobachten. Im Vergleich dazu konnte die GJ366665U-Variante nicht mit einem Einfluss auf die Virusvermehrung in Verbindung gebracht werden. Nichtsdestotrotz, konnte die GJ366665U-Variante, und damit eine erhöhte Expression von G, mit einer gesteigerten Virulenz assoziiert werden, während die GJ366665A-Variante, d. h. eine verringerte G-Expression zur Attenuierung des Virus führte. Die Untersuchungen wurden auf weitere G-Genstrukturen, d.h. ein rekombinantes PVM, rPVM-Gt, das ein N-terminal verkürztes G-Protein exprimiert, ausgeweitet. Zum Vergleich wurde eine Deletionsmutante des kompletten G-Gens, rPVM-ΔG, mit einbezogen. Von beiden Viren war bereits bekannt, dass sie in vivo attenuiert sind. Die Organisation des Gt-Gens war mit einer um 75 % verringerten Expression des entsprechenden Proteins assoziiert, was zuvor nicht beobachtet worden war. Zugleich zeigte rPVM-Gt eine deutliche Attenuierung der Replikation in RAW 264.7-Zellen und primären Mausmakrophagen, die von der G-Deletionsmutante noch übertroffen wurde. Die im Vergleich zu der GJ366665A-Variante deutlich höhere Reduktion der G-Expression dieser beiden G-Mutanten in Betracht ziehend, scheint dies darauf hinzuweisen, dass eine bestimmte Mindestexpression von G für eine effiziente Virusvermehrung in diesen Zellen benötigt wird. Zusammenfassend deuten die erhaltenen Ergebnisse darauf hin, dass die Expression des G-Proteins durch die jeweiligen 5’ nicht-translatierte Region des Gens moduliert wird, was einen neuen Mechanismus für Negativstrang-RNA-Viren darstellt. Zugleich moduliert die Expressionsrate von G die Virulenz von PVM.

No description available.

Increasing evidence suggests that overlapping genes is a common phenomenon in eukaryotic genomes too and are not restricted to prokaryotes alone. Here we determined overlapping genes in a set of orthologous genes in the genomes of human, chimp, mouse, rat, and dog and contrasted the patterns of overlapping between two principal types of overlapping genes, the same-strand-overlapping genes and different-strand-overlapping genes. The two types of overlapping genes are compared with respect to their frequencies, overlap lengths, region of overlap, and conservation of overlap in five species. Our results suggest the following: different-strand-overlaps are more common, both types show different patterns with respect to overlap lengths and regions of overlap, different-strand-overlapping genes are more evolutionarily conserved, and 3'-UTR evolution plays an important role in transitions between non-overlapping genes and overlapping genes. The thesis also presents a review of related work in terms of history, origin, types, biological significance of overlapping genes, human diseases associated with them, and their comparison in prokaryotes and eukaryotes. / Master of Science

Introdução: A síndrome Richieri-Costa-Pereira (SRCP; OMIM 268305) é uma disostose acrofacial autossômica recessiva caracterizada por fissura mandibular mediana associada a outros defeitos craniofaciais e de membros. Objetivo: Mapear o gene causador da SRCP. Métodos e Resultados: O gene da doença foi mapeado para uma região de 128.5kb no cromossomo 17q25.3, por meio de análise de identidade por descendência, em 22 genealogias. As estratégias de sequenciamento identificaram ampliação de motivos com 18 ou 20 nucleotídeos da região 5\' não traduzida (5\' UTR) do gene EIF4A3. Observou-se 15 ou 16 repetições dos motivos nos indivíduos afetados e de 3 a 12 repetições nos indivíduos do grupo controle (520 indivíduos). Estudo de expressão detectou níveis de transcrição do gene EIF4A3 acentuadamente reduzidos nas células brancas do sangue de um dos indivíduos com SRCP, quando comparado aos cinco controles testados. Estudo funcional por meio de modelamento do ortólogo eif4a3 em zebrafish resultou no subdesenvolvimento de estruturas cartilaginosas e ósseas craniofaciais derivadas do primeiro ao quinto arcos faríngeos, similar às alterações craniofaciais observadas nos indivíduos com SRCP. Conclusão: Estes estudos confirmam que a SRCP é causada por deficiência de EIF4A3 e, ainda, que esse gene é essencial para a morfogênese da mandíbula, da laringe e dos membros, importantes estruturas comprometidas na SRCP. / Introduction: Richieri-Costa-Pereira syndrome (RCPS; MIM#268305) is an autosomal recessive acrofacial dysostosis characterized by mandibular median cleft associated with other craniofacial and limb defects. Purpose: Map the gene that causes the SRCP. Methods and Results: We mapped the disease gene to a 128.5kb region on chromosome 17q25.3 through identity-by-descent analysis in 22 genealogies. Sequencing strategies identified an expansion of a complex region with several repeats of 18 or 20-nucleotide motifs in the 5 untranslated region (5´UTR) of EIF4A3, which contained from 15 or 16 repeats in the affected patients and from 3 to 12 repeats in 520 healthy individuals. EIF4A3 transcriptional levels were markedly reduced in white blood cells of a RCPS patient as compared to controls. Furthermore, modeling of the ortholog eif4a3 in zebrafish lead to underdevelopment of several cartilage and bone craniofacial structures derived from the 1st to 5th pharyngeal arches, in agreement with the craniofacial alterations of RCPS. Conclusion: These studies thus confirm that RCPS is caused by deficiency of EIF4A3, and assign EIF4A3 for the first time as essential for mandibular and also for larynx and limb morphogenesis, important structures impaired in RCPS.

Submitted by Ana Maria Fiscina Sampaio (fiscina@bahia.fiocruz.br) on 2012-07-25T21:04:26Z No. of bitstreams: 1 Marcela Kelly Astete Gómez Estudo dos alelos da região....pdf: 1187772 bytes, checksum: d35081c0e79dae64122fa5e5a15c60c3 (MD5) / Made available in DSpace on 2012-07-25T21:04:26Z (GMT). No. of bitstreams: 1 Marcela Kelly Astete Gómez Estudo dos alelos da região....pdf: 1187772 bytes, checksum: d35081c0e79dae64122fa5e5a15c60c3 (MD5) Previous issue date: 2011 / Fundação Oswaldo Cruz. Centro de Pesquisas Gonçalo Moniz. Salvador, Bahia, Brasil / A Síndrome do X-Frágil (SXF) é a principal causa hereditária de deficiência mental de herança dominante ligada ao cromossomo X. Em 1991, o gene FMR1 (Fragile X Mental Retardation 1) foi descoberto como responsável pela SXF. De acordo com o número de repetições, os alelos se dividem em: (a) alelo normal compreendendo 6 a 55 repetições (b) pré-mutado: 61 a 200 repetições e (c) mutado: com >200 repetições. Os alelos que apresentam 45 a 60 repetições são considerados como zona intermediária ou gray-zone. A base molecular dessa doença é bastante peculiar quando comparada com os padrões típicos observados em outras desordens de etiologia genética. As pré-mutações podem agregar-se de forma silenciosa por muitas gerações de uma família antes de se expandir para a mutação completa, levando aos sinais clínicos da doença. Na Bahia existem poucas pesquisas envolvendo esta doença, consequentemente, é desconhecida sua frequência nessa região. Portanto, este trabalho tem como objetivo detectar a frequência das classes alélicas do gene FMR1 em uma amostra de indivíduos do sexo masculino da população geral de Salvador-BA. Este estudo foi realizado em 511 homens provenientes de outro amplo estudo desenvolvido pelo Instituto de Saúde Coletiva (ISC) da Universidade Federal da Bahia (UFBA) sendo utilizada a técnica da PCR e posterior classificação dos alelos baseando-se na relação entre o número de repetições CGG e o tamanho do fragmento obtido na PCR. Dos 511 homens analisados no presente estudo observou-se predominância de alelos normais pertencentes à classe 2 (11 a 26 repetições CGG) em 73,70% dos indivíduos analisados, seguido da classe 3 (27 a 40 repetições CGG) em 25,10% dos indivíduos. Apenas 1,20% foram incluídos na classe 1 (<10 repetições CGG) e nenhum alelo foi encontrado nas classes 4 (41 a 60 repetições CGG), classe 5 ( >60 repetições CGG) e classe 6 (>200 repetições CGG). Este é o primeiro estudo utilizando a técnica da PCR para detecção dos alelos do gene FMR1 em uma população geral de Salvador-BA, podendo direcionar futuros trabalhos envolvendo o gene FMR1 tanto para o estado da Bahia, quanto para a região Nordeste e também minimizar deficiências existentes em termos de diagnóstico da SXF na cidade de Salvador-BA. / The Fragile X syndrome (FXS) is the leading cause of inherited mental deficiency (MD) of dominant inheritance linked to X chromosome. In 1991, the FMR1 gene (Fragile X Mental Retardation 1) was discovered as responsible for FXS. According to the number of repetitions, the alleles are divided in: (a) normal stable allele comprising 6 to 55 repetitions; (b) premutation: 61 to 200 repetitions and (c) mutant with> 200 repeats. The molecular basis of this disease is quite unusual when compared with the typical patterns seen in other disorders of genetic etiology. The pre-change can add up so silent for many generations of a family before they expand to full mutation, leading to clinical signs of disease. In Bahia there are few studies involving this disease, therefore, its frequency is unknown in this region. So, this study aims to detect the frequency of allelic classes of FMR1 gene in a sample of males from the general population of Salvador-BA city. This study was conducted on 511 samples using the PCR technique and subsequent classification of alleles based on the number of CGG repeats and the size of the fragments in PCR. Of the 511 individuals examined was found to predominate among the normal alleles the class 2 (11 to 26 repetitions) with 73,70% followed alleles analyzed in Class 3 (27 to 40 repetitions) with 25,10%, and Class 1 (<10 repetitions) with only one 1,20% . No allele was found in class 4 (41 to 60 repetitions), which corresponds to gray zone and class 5 (> 60 repetitions) for the premutation and class 6 (>200 repetitions) the corresponds full mutation.This is the first study using the PCR technique for detection of alleles of the FMR1 gene in a general population of Salvador, Bahia, and may direct future studies involving the FMR1 gene for both the state of Bahia, the Northeast and to also minimize deficiencies existing in the diagnosis of FXS in Salvador, Bahia.

Introdução: A síndrome Richieri-Costa-Pereira (SRCP; OMIM 268305) é uma disostose acrofacial autossômica recessiva caracterizada por fissura mandibular mediana associada a outros defeitos craniofaciais e de membros. Objetivo: Mapear o gene causador da SRCP. Métodos e Resultados: O gene da doença foi mapeado para uma região de 128.5kb no cromossomo 17q25.3, por meio de análise de identidade por descendência, em 22 genealogias. As estratégias de sequenciamento identificaram ampliação de motivos com 18 ou 20 nucleotídeos da região 5\' não traduzida (5\' UTR) do gene EIF4A3. Observou-se 15 ou 16 repetições dos motivos nos indivíduos afetados e de 3 a 12 repetições nos indivíduos do grupo controle (520 indivíduos). Estudo de expressão detectou níveis de transcrição do gene EIF4A3 acentuadamente reduzidos nas células brancas do sangue de um dos indivíduos com SRCP, quando comparado aos cinco controles testados. Estudo funcional por meio de modelamento do ortólogo eif4a3 em zebrafish resultou no subdesenvolvimento de estruturas cartilaginosas e ósseas craniofaciais derivadas do primeiro ao quinto arcos faríngeos, similar às alterações craniofaciais observadas nos indivíduos com SRCP. Conclusão: Estes estudos confirmam que a SRCP é causada por deficiência de EIF4A3 e, ainda, que esse gene é essencial para a morfogênese da mandíbula, da laringe e dos membros, importantes estruturas comprometidas na SRCP. / Introduction: Richieri-Costa-Pereira syndrome (RCPS; MIM#268305) is an autosomal recessive acrofacial dysostosis characterized by mandibular median cleft associated with other craniofacial and limb defects. Purpose: Map the gene that causes the SRCP. Methods and Results: We mapped the disease gene to a 128.5kb region on chromosome 17q25.3 through identity-by-descent analysis in 22 genealogies. Sequencing strategies identified an expansion of a complex region with several repeats of 18 or 20-nucleotide motifs in the 5 untranslated region (5´UTR) of EIF4A3, which contained from 15 or 16 repeats in the affected patients and from 3 to 12 repeats in 520 healthy individuals. EIF4A3 transcriptional levels were markedly reduced in white blood cells of a RCPS patient as compared to controls. Furthermore, modeling of the ortholog eif4a3 in zebrafish lead to underdevelopment of several cartilage and bone craniofacial structures derived from the 1st to 5th pharyngeal arches, in agreement with the craniofacial alterations of RCPS. Conclusion: These studies thus confirm that RCPS is caused by deficiency of EIF4A3, and assign EIF4A3 for the first time as essential for mandibular and also for larynx and limb morphogenesis, important structures impaired in RCPS.

Les plantes sont ancrées au sol pendant la majorité de leur cycle de vie et doivent donc constamment adapter leur croissance et leur métabolisme aux stress abiotiques. Ainsi, la subsistance des plantes dépend de leur capacité à réguler rapidement l’expression des gènes afin d’adapter leur physiologie à l’environnement. L’expression d’un gène peut être contrôlé à plusieurs niveaux; transcriptionnel, post-transcriptionnel, traductionnel et post-traductionnel.De nombreux processus cellulaires vitaux tels que la réplication de l’ADN, la transcription, la synthèse protéique, et la dégradation des protéines, sont régulés par les signaux environnementaux. Des études chez la levure, la drosophile et les animaux ont montré que la protéine kinase TOR (Target Of Rapamycin) est impliquée dans le contrôle de la croissance cellulaire et de la prolifération en réponse à différents signaux tels que les nutriments, les acides aminés, les hormones et les facteurs de croissance. Chez Arabidopsis thaliana, TOR est nécessaire au développement de l’embryon et de l’endosperme. De plus, des modifications du niveau de protéine AtTOR affectent la croissance végétative et la reproduction.Le principal objectif de cette thèse est de caractériser les mécanismes qui contrôlent l’expression de AtTOR en déterminant les éléments de régulation situés sur le la région 5’ non traduite (5′UTR) de l’ARNm de AtTOR, puis de manipuler ces éléments de régulation afin d étudier leur rôle. Nous avons choisi de nous focaliser sur la région 5′UTR de AtTOR, et sur une microORF (uORF) située en amont de l’ORF principale de AtTOR. Il s’agit de la première tentative d’étude de la régulation de l’expression de TOR par ces éléments chez les eucaryotes.Trois constructions chimériques ont été réalisées pour cette étude et transformée transitoirement est de manière stable dans des plantes. La première construction (contrôle positif) incluse le promoteur de AtTOR, la région 5′UTR, le premier intron et le début du premier exon fusionné au gène rapporteur GUS. La seconde construction (microORF mutée) est présente une mutation du codon start de la microORF (ATG changé en TTG). Enfin, la troisième construction (5′UTR délétée) contient la même séquence que le contrôle positif mais sans la région 5′UTR. Ces constructions ont également été placée sous le contrôle du promoteur 35S au lieu du promoteur de AtTOR afin d’étudier un lien éventuel entre la 5′UTR et la microRF et le promoteur de AtTORNos résultats indiquent une régulation généralement négative exercée par la 5′UTR, et dans une moindre mesure par la microORF, sur l’expression de AtTOR. Cette régulation semble avoir lieu au niveau transcriptionnel ou au niveau de la stabilité de l’ARNm, mais pas au niveau de la traduction. En effet, les modifications du niveau de transcrit GUS sont suivie d’un changement équivalent de l’activité GUS. De plus, nous avons observé que l’auxine et le sucrose ont un effet positif sur l’expression de AtTOR. Dans le cas de l’auxine, cet effet semble lié à la présence de la région 5′UTR de AtTOR.D’autres études de la fonction de la région 5’UTR et de la microORF de AtTOR, ainsi que de leur relation avec d’autres éléments régulateurs localisée dans le promoteur de AtTOR, permettront de mieux comprendre comment ces éléments régulateurs contrôlent finement l’expression de AtTOR. / Land plants are anchored in one place for most of their life cycle and therefore must constantly adapt their growth and metabolism to abiotic stresses. Thus, plants’ subsistence depends on their ability to regulate rapidly gene expression in order to adapt their physiology to their environment. The expression of a gene can be controlled at many levels, including transcription, post-transcription, translation, and post-translation.Many vital cellular processes like DNA replication, transcription, protein synthesis, and protein degradation are regulated by environmental signals. Studies in yeast, Drosophila, and mammals showed that the target of rapamycin (TOR) protein is involved in control of cell growth and cell proliferation in response to different types of environmental signals such as nutrients, amino acids, hormones, and growth factors. In Arabidopsis thaliana, TOR is necessary for both embryo and endosperm development in, and changes of TOR protein level affect both vegetative and reproductive growth.The main purpose from this thesis is to highlight the mechanisms that control AtTOR expression at the post-transcriptional level through determination of the possible regulatory elements within the 5′ untranslated region (5′UTR) or the first intron of AtTOR mRNA itself, and through manipulation of these regulatory elements to study their precise role. We have chosen to focus on the small upstream open reading frame (uORF) as well as the 5′UTR region. This is the first attempt to study the regulation of TOR kinase expression in eukaryotes through these small uORF or the sequence of 5′ untranslated region (5′UTR).To achieve this purpose, three chimeric constructs have been established and transformed in Nicotiana benthamiana leaves and Arabidopsis thaliana plants. The first construct (the positive control) contains the AtTOR promoter, the 5′UTR, the first intron, and the beginning of the second exon fused to the GUS reporter gene. The second construct (mutated uORF) have the same sequence as the positive control construct except the start codon of uORF was changed from ATG to TTG. The third construct (deleted 5′UTR) have the same sequence as the positive control construct without the 5′UTR. These constructs have also been placed under the activity of CaMV 35S promoter instead of AtTOR promoter to investigate whether there is a link between the 5′UTR/or uORF and the promoter.Our work show an overall negative regulation exerted by the 5′UTR and, to a lesser extent, by the uORF on AtTOR gene regulation. This regulation is likely at the level of transcription or mRNA stability, since the changes in GUS transcript level was followed by the same changes in GUS activity. In addition we found that external inducers like auxin or sucrose exert a positive effect on AtTOR expression. This effect appears somehow linked to the presence of the 5′UTR of AtTOR mRNA.Greater insight into the molecular mechanisms of AtTOR 5′UTR/or uORF function and its relationship with other regulatory elements located in AtTOR promoter will be required to understand how these regulatory elements work either individually or in combination to achieve the fine and accurate regulation of their gene expression.