La TriméthylGuanosine Synthase 1 de levure (Tgs1) à été identifiée à la suite d’un criblage double hybride en utilisant l’extrémité basique carboxy-terminale de la protéine SmB, cœur des snRNP, comme appât. Il a été également montré que Tgs1 interagit spécifiquement avec le domaine carboxyl-terminal basique KKD/E des protéines Nop58p et Cbf5p, deux composants protéiques du coeur des snoRNP. Le gène TGS1 n’est pas essentiel mais sa délétion confère un phénotype de cryo-sensibilité associé à un léger défaut d’épissage à basse température, associé à la rétention de U1 dans le nucléole. La recherche de substrats pour cette protéine a montré que Tgs1p est capable de méthyler la coiffe monométhylée des snARN et des snoARN transcrits par l’ARN polymérase II. La grande majorité des snoARN joue un rôle dans la sélection des sites de modifications de plusieurs classes d’ARN. Certains, par contre, sont impliqués dans la voie de synthèse des ribosomes, un processus comprenant de multiples étapes de clivages endo- et exoribonucléotidiques et ayant lieu dans le nucléole où les facteurs impliqués dans ces réactions se concentrent en plusieurs domaines distincts. Le point de départ de ce travail de thèse a été de tester un possible rôle de Tgs1p et/ou de la triméthylation dans la biosynthèse du ribosome. Dans un premier temps, l’analyse du processing des ARN ribosomiques dans la souche délétée pour TGS1 nous a permis de mettre en évidence l’implication de Tgs1 dans la formation de l’ARNr de la petite sous-unité, l’ARNr 18S. Des mutants catalytiques de Tgs1, incapables de reconnaître et de modifier les coiffes m7G, ont été crées. L’analyse de la voie de biogenèse des ribosomes dans ces souches ne présente pas les défauts constatés dans la souche délétée, révélant que c’est la protéine et non sa fonction catalytique qui est requise. De plus, ces mutants sont autant défectueux dans l’épissage des ARN messagers, excluant toute implication du défaut d’épissage dans le ralentissement de la voie de biogenèse des ribosomes observé dans la souche délétée. L’ultrastructure des souches délétées pour TGS1 observée en microscopie électronique nous a permis de mettre en évidence un effet de l’absence de Tgs1 sur la morphologie nucléolaire. En effet, le nucléole dans ces souches ne présente plus de nucléole structuré, bi-compartimenté. Les analyses en microscopie à fluorescence ont confirmé la disparition de la ségrégation des deux compartiments nucléolaires, suggérant que le défaut dans la biogenèse des ribosomes puisse être une conséquence de la perte de cohérence du nucléole. La caractérisation de l’environnement physique et fonctionnel de Tgs1 a été entreprise afin de mettre à jour des fonctions additionnelles de la protéine. Diverses approches ont été envisagées: la recherche de partenaires physiques par l’emploi d’un allèle de TGS1 étiquetté TAP permettant la purification puis l’analyse de partenaires physiques ainsi que la recherche de partenaires fonctionnels par la méthode du crible synthétique létal. La recherche de partenaires physiques a permis de révéler l’existence d’un grand nombre d’ARN non codants coprécipités avec Tgs1. Certains sont des substrats connus de la protéine mais un grand nombre d’ARN ne possédant pas de coiffes monométhylées. La recherche de partenaires fonctionnels a permis la découverte de candidats synthétiques létaux appartenant à deux groupes, un groupe lié à l’épissage des ARN messagers et un autre groupe constitué de membres du complexe SWR1, complexe impliqué dans la régulation transcriptionnelle par modification de la chromatine. Lors de ce crible de candidats synthétiques létaux, il est apparu que la délétion de TGS1 restaure partiellement le défaut de croissance à chaud induit par la délétion du gène RRP47, dont le produit est impliqué dans la maturation de l’extrémité 3’ de plusieurs types d’ARN non codants. Les travaux préliminaires effectués ne permettent pas encore d’expliquer un tel phénotype. Au cours de ce travail de thèse, nous avons pu répondre à un certain nombre de questions sur la fonction et le rôle de Tgs1 dans la cellule. La fonction catalytique de Tgs1 dans la méthylation des coiffes m7G est clairement nécessaire à l’efficacité de l’épissage des ARN messagers mais le rôle de la triméthylation de la coiffe des snoARN n’est pas élucidé à ce jour. Le fait que la fonction catalytique de Tgs1 n’est pas impliquée dans le défaut dans la biogenèse des ribosomes et la découverte du rôle de la protéine dans la morphologie nucléolaire, laisse entrevoir l’existence de fonctions additionnelles de Tgs1 dans la cellule. La caractérisation de son environnement physique et fonctionnel abonde justement dans ce sens, mettant à jour plusieurs interactions probablement liées à sa fonction catalytique, notamment dans l’épissage des ARN messagers mais également un grand nombre d’interactions impliquant la participation de Tgs1 dans d’autres voies métaboliques.

<p>Bisphosphate 3'-nucleotidase (Bpnt1) is a member of a family of small molecule phosphatases whose activities depend on divalent cations and are inhibited by lithium. While the enzymes share many commonalities, they have distinct and non-­overlapping substrate pools. Of the seven mammalian members, two enzymes, gPAPP and Bpnt1, hydrolyze the same small molecule 3'-phosphoadenosine 5'-phosphate (PAP) but act in separate subcellular compartments, the Golgi apparatus and cytoplasm respectively. Hydrolysis of PAP, which is a metabolite of the inorganic sulfate incorporation pathway, is highly conserved throughout evolution from bacteria to yeast to humans. Evidence in multiple species has shown that inhibiting PAP hydrolysis leads to cellular toxicity as a result of its accumulation and also that these effects can be ameliorated by modulating the rate of its production. However, despite the abundant evidence of its importance</p><p>from studies in lower eukaryotes, the role of the cytoplasmic PAP phosphatase, Bpnt1, in more complicated mammalian physiological remains poorly understood. Here we report for the first time the generation and characterization of mice deficient for Bpnt1. Bpnt1 null mice do not exhibit skeletal defects, but instead develop severe liver</p><p>pathologies and deficiencies in intestinal iron absorption. Loss of Bpnt1 leads to tissue-specific elevations of the substrate PAP. To test the hypothesis that a toxic cellular accumulation of PAP accounts for the observed phenotypes, we generated a double mutant mouse that concomitantly down regulates bisphosphorylated nucleotide synthesis in the context of Bpnt1 deficiency. Remarkably, double mutants do not display any detectable physiological defects seen in Bpnt1 null mice. In addition, we have identified and characterized a novel substrate of 3'­nucleotidases, 3'-phosphoadenosine 5'-diphosphate (PAPP) that co-accumulates with PAPS and PAP and might play a role in mediating certain aspects of the physiological defects of Bpnt1 null mice. Overall, our study defines a role for Bpnt1 in mammalian physiology and provides mechanistic insights into the importance of cytoplasmic 3'-­</p><p>nucleotide hydrolysis to normal cellular function.</p> / Dissertation

Radial chromosome positioning in interphase nuclei is non-random and can alter according to developmental, differentiation, proliferation or disease status. The aim of this thesis is to understand how chromosome re-positioning is elicited and to identify the nuclear structures that assist this re-localisation event. By positioning all human chromosomes in primary fibroblasts that have left the proliferative cell cycle, the study within this thesis has demonstrated that in cells made quiescent by reversible growth arrest, chromosome positioning is altered considerably. Upon removal of serum from the culture medium, chromosome re-positioning took less than 15 minutes, required energy and was inhibited by drugs affecting the polymerization of myosin and actin. The nuclear distribution of nuclear myosin 1β was dramatically different in quiescent cells as compared to proliferating cells. If the expression of nuclear myosin 1β was suppressed using interference RNA procedures the movement of chromosomes after 15 minutes in low serum was inhibited. When high serum was restored to the serum starved cultures chromosome repositioning was only evident after 24-36 hours that coincided with a return to a proliferating distribution of nuclear myosin 1β.

The nucleolus is the site for rRNA synthesis, a process requiring the recruitment of many proteins involved in ribosomal biogenesis. Nrap is a novel nucleolar protein found to be present in all eukaryotes. Preliminary characterisation of Nrap suggested it was likely to participate in ribosome biogenesis but as with many other nucleolar proteins, the functional role of Nrap is largely unknown. In this study, the role of mammalian Nrap in the nucleolus and in ribosome biogenesis was explored. Initially, a number of tools were generated to investigate Nrap function. This involved raising and purifying a polyclonal antibody against the N-terminal region of Nrap. The anti-Nrap antibody was found to detect two Nrap bands in mouse fibroblast cells, possibly corresponding to the two mouse Nrap isoforms, and . In addition, mammalian expression vectors containing the full Nrap sequence as well as deletion constructs were created. The subcellular localisation of each construct was observed by fluorescent microscopy. It was revealed that recombinant Nrap did not localise to the nucleolus, possibly because it was exported to undergo degradation by the 26S proteasome. Two putative NLSs were found to be responsible for directing Nrap to the nucleus but a region accountable for nucleolar localisation was not identified. The data indicated that multiple domains working together are likely to direct Nrap to the nucleolus. Nrap was also observed to co-localise with nucleolar proteins B23 and p19ARF. Moreover, it was shown by reciprocal immunoprecipitation that these three nucleolar proteins existed in a complex in unsynchronised mouse fibroblast cells. Recent reports demonstrated a complex relationship between B23 and p19ARF although the functional significance remained unclear. Nrap's in vivo association with B23 and p19ARF indicated a specific functional role in the nucleolus. Nrap knockdown using siRNA significantly increased B23 protein levels in a dose-dependent manner and down-regulated p19ARF protein levels at higher siRNA concentration. Preliminary studies also implicated Nrap in cell proliferation through these novel interactions. Both endogenous and recombinant Nrap were found to be highly unstable suggesting that Nrap might regulate B23 and p19ARF through its own tightly regulated stability. Finally, the role of Nrap in rRNA processing was investigated by northern blot analysis. Nrap knockdown was found to affect the levels of 45S, 32S and 28S rRNAs. The changes found may be a consequence of the concurrent perturbation in the levels of B23 and p19ARF caused by Nrap knockdown. As the results were not consistent with previous reports, it was likely that changes to rRNA processing could be contributed to Nrap loss of function. This study demonstrated for the first time a functional role of Nrap in rRNA processing possibly through its association with B23 and p19ARF.

The nucleolus of eukaryotic cells contain a number of the intron-coding small nucleolar RNAs (snoRNAs), which functions are related to covalent modification of pre-rRNAs. The snoRNA that from long, phylogenetically conserved sequence complementarity to 28S, 18S, 5.8S and 5S rRNAs are designated as 28S, 18S, 5.8S and 5S snoRNAs, respectively. In the present study, studying on human 5S snoRNAs had been carried out. The human genome encoding candidate 5S snoRNAs were searched using database mining. The transcripts of 5S snoRNA genes were identified by RT-PCR analyses and DNA sequencing. No appreciable diversities of 5S snoRNA genes were observed as evidenced by single strand conformation polymorphism (SSCP) and high resolution agarose gel. Moreover, sequence conservation of 5S snoRNAs reflects a requirement for maintaining their secondary structure on exerting their function. The results of RT-PCR analyses revealed a tissue-specific transcription of 5S snoRNAs. A 5S snoRNA designated as N117 was identified to be highly expressed in normal brain. On the contrary, its expression markly decreased in brain tumor (meningioma). This seems to be associated with the expression of host gene, which encodes a protein similar to synapsin III protein. Consequently, this may implicate that the use of snoRNA as a potential index for the transcription of its host gene.

Diss. (sammanfattning) Stockholm : Karol. inst., 2003. / Härtill 5 uppsatser.

Ribosomes are macromolecular machines that are responsible for production of every protein in a living cell. Yet we do not know the details about how these machines are formed. The ribosome consists of four RNA strands and roughly 80 proteins that associate with each other in the nucleolus and form pre-ribosomal complexes. Eukaryotes, in contrast to prokaryotes, need more than 200 non-ribosomal factors to assemble ribosomes. These associate with pre-ribosomal complexes at different stages as they travel from the nucleolus to the cytoplasm and are required for pre-rRNA processing. We do however lack knowledge about the molecular function of most of these factors and what enables pre-rRNA processing. Especially, information is missing about how non-ribosomal factors influence folding of the pre-rRNA and to what extent the pre-ribosomal complexes are restructured during their maturation.  This thesis aims to obtain a better understanding of the earliest events of ribosome assembly, namely those that take place in the nucleolus. This has been achieved by studying the essential protein Mrd1 by mutational analysis in the yeast Saccharomyces cerevisiae as well as by obtaining structural information of nucleolar pre-ribosomal complexes. Mrd1 has a modular structure consisting of multiple RNA binding domains (RBDs) that we find is conserved throughout eukarya. We show that an evolutionary conserved linker region of Mrd1 is crucial for function of the protein and likely forms an essential module together with adjacent RBDs. By obtaining structural information of pre-ribosomal complexes at different stages, we elucidate what structuring events occur in the nucleolus.  We uncover a direct role of Mrd1 in structuring the pre-rRNA in early pre-ribosomal complexes, which provides an explanation for why pre-rRNA cannot be processed in Mrd1 mutants. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 3: Manuscript. Paper 4: Manuscript.</p>

Defective chromatin remodeling proteins are associated with both germline and acquired human disease. PHF6 is encoded by an X-linked gene that is predominantly expressed in the brain and thymus. Structurally, PHF6 contains nuclear and nucleolar localization sequences as well as two ZaP domains, which bind dsDNA. Germline mutations in PHF6 are the cause of BFLS, an XLID, while somatic PHF6 mutations have been identified in T-ALL, AML, and CML. Indeed, screening of a pediatric cohort of nine T-ALL patients revealed a novel H329Q mutation. In a further clinical analysis, T-ALL onset occurred in a 9-year old male BFLS patient with an R342X mutation, suggesting that BFLS might be a cancer predisposition syndrome. To better understand its protein function, recombinant PHF6 was co-immunoprecipitated for a mass spectrometry based proteomic screen. Notably, PHF6 co-purified with multiple constituents of the NuRD complex, an important transcriptional regulator during embryogenesis and lineage commitment with particularly well characterized responsibilities during lymphogenesis. PHF6-NuRD localization was restricted to the nucleoplasm, however PHF6 also co-purified with several ribosomal and splicing proteins. When examined further, PHF6 was found to be recruited to the nucleolus by an RNA-mediated interaction and co-localized within the subnucleolar FC and DFC compartments. ChIP-qPCR revealed that PHF6 binds to transcribed regions of rDNA, resulting in the repression of rRNA. These data thus present a model of PHF6 acting as a tumour suppressor by mediating both nucleoplasmic and nucleolar transcriptional events.

Is the nucleolus a sine qua non condition of eukaryotes? The present project starts from this central question to contribute to our knowledge about the origin and the evolution of the cells. The nucleolus is a cryptic organelle that plays a central role in cell function. It is responsible for the orchestration of ribosomal RNA expression, maturation and modification in the regulatory context of cellular homeostasis. Ribosomal expression is undoubtedly the greatest transcriptional and regulatory activity of any cell. The nucleolus is not just a conventional organelle –membrane-limited-, but a magnificent transcriptional puff: a dichotomy between structure and process, form and function. What is the minimum nucleolus? Evolution should bring some light into these questions. Evolutionary cell biology (ECB) has raised increasing attention in the last decades. Is this a new discipline and an historical opportunity to combine functional and evolutionary biology towards the insight that cell evolution underlies organismic complexity? In the post-genomic era, we have developed the potential of combining high throughput acquisition of data with functional in situ and in sillico approaches: integration understood as omics approaches. Can this provide a real consilience between evolutionary and functional approaches? The reduced number of model organisms and cultivation techniques still excludes the majority of the extant diversity of cells from the scope of experimental inquiry. Furthermore, at the conceptual level, the simplification of evolutionary processes in biosciences still limits the conformation of a successful disciplinary link between functional and evolutionary biology. This limits the formulation of questions and experiments that properly address the mechanistic nature of cellular events that underlie microbial and organismic diversity and evolution. Here we provide a critical and comparative review to the historical background of ECB. This project takes the lessons learned from ECB and attempts to find a homologue structure of the eukaryotic nucleolus within the Archaea. We found nanometric structures in S. solfactarius that either are positive to specific nucleolar techniques such as Nucleolar organizer regions NOR silver staining. These is structures are novel and its significance should be revised on the evolutionary cell biology perspective.

Selenoprotein H (SELENOH) is a nucleolar oxidoreductase with DNA binding properties whose function is not well understood. To determine the functional and physiological roles of SELENOH, a knockout of SELENOH was generated in cell lines using CRISPR/Cas9-mediated genomic deletion and in mice by targeted disruption. Based on the sequenced genome, the results of deduced protein sequences indicated various forms of mutants in the CRISPR/Cas9-mediated knockout, including a frame-shift by aberrant splicing and truncated SELENOH by early termination of the translation process. Loss of SELENOH in HeLa cells induced slow cell proliferation, the formation of giant multinucleated cells, accumulation of unrepaired DNA damage and oxidative stress, and cellular senescence. SELENOH cells were enlarged and possessed a single large nucleolus. Atomic force microscope showed increased stiffness in the nucleoli of SELENOH knockout cells, which suggests that SELENOH maintains the flexible structure of the nucleolus. Furthermore, the knockout of SELENOH led to a large-scale reorganization of the nucleolar architecture with the movement of nucleolar protein into nucleolar cap regions in response to oxidative stress. The nucleolar reorganization is dependent on ATM signaling. Altogether, results suggest that SELENOH appears to be a sensor of oxidative stress that plays critical roles in redox regulation and genome maintenance within the nucleolus. To determine the physiological role of SELENOH in vivo, Selenoh knockout mice were generated by targeted deletion through homologous recombination. Selenoh+/− mice were fertile and phenotypically indistinguishable from wild-type littermates. Results from matings of Selenoh+/− mice showed a significantly reduced fraction of Selenoh−/− offspring on the basis of Mendelian segregation. Since some Selenoh−/− were born, it is likely that Selenoh is a partially essential gene in mice. Live-born Selenoh−/− mice were viable and born without apparent phenotypes. Selenoh−/− mice at 2-month of age showed increased GPX activity in the lung but not in the brain and liver. Furthermore, loss of Selenoh resulted in the aggravated formation of aberrant crypt foci in the colon of Selenoh+/− mice that were injected with azoxymethane. Altogether, SELENOH has critical roles in embryogenesis and colorectal carcinogenesis.