DNA methylation patterns were thought to be stable and unchanging throughout life. However more evidence has emerged suggesting that the DNA methylation pattern is dynamic, subject to changes in response to physiological and environmental cues. Thus, it is crucial to understand when these changes occur and how the DNA methylation pattern can be manipulated. This thesis focuses on alterations in the DNA methylation pattern during the cell cycle both globally and at a specific promoter, as well as how the methylated DNA binding domain protein MBD3 is involved in regulating these changes. During the cell cycle, the levels of DNA methylation are highest during S phase, and drop after DNA replication during G0/G1. These changes in DNA methylation are not occurring at repetitive sequences, but rather at specific sequences throughout the genome, as determined using a CpG island microarray. Conversely, the DNA methylation pattern of the ribosomal RNA (rRNA) promoter follows an opposite pattern, with decreased levels of DNA methylation at G1/S and S phase, with the levels increasing as the cell progresses through the cell cycle. These changes at the rRNA promoter correlate with other epigenetic changes, including histone acetylation and transcription factor binding, indicating a coordinated epigenetic change regulated with the different stages of the cell cycle. We further determined how the DNA methylation pattern of the rRNA promoter is regulated, and found a critical role for the methylated DNA binding domain protein MBD3. MBD3 was found to bind to unmethylated rRNA promoters, together with the transcription factor Upstream Binding Factor (UBF). Manipulating the levels of MBD3 expression resulted in a change in the levels of DNA meth / Le patron de méthylation de l'ADN est connu pour être stable et fixes durant toute la vie. Cependant, plusieurs évidences suggèrent que le patron de méthylation de l'ADN soit dynamique, sujet aux changements en réponse à des stimuli physiologiques et environnementaux. Ainsi, il est crucial de comprendre quand ces changements se produisent et comment le patron de méthylation d'ADN peut être manipulé. Cette thèse se concentre sur les changements dans le patron de méthylation de l'ADN pendant le cycle cellulaire, à l'échelle globale et au niveau de promoteurs spécifiques, et également sur le mécanisme par lequel la protéine MBD3, qui contient un domaine de liaison à l'ADN méthylé, module ces changements. Pendant le cycle cellulaire, les niveaux de méthylation de l'ADN sont plus élevés pendant la phase de S, et chute par la suite après la réplication de l'ADN lors de la phase G0/G1. Ces changements de méthylation de l'ADN ne se produisent pas dans des séquen-ces répétitives, mais plutôt dans des séquences spécifiques dans tout le génome, tel que déterminé par microarray des îlots CpG. Réciproquement, le patron de méthylation du promoteur de l'ARN ribosomal (ARNr) suit un modèle opposé, avec un bas niveau de méthylation de l'ADN en phase G1/S et S, qui augmente progressivement pendant le cycle cellulaire. Ces changements dans le promoteur de l'ARNr corrèlent avec d'autres changements épigénétiques, y compris l'acétylation des histones et la liaison de facteurs de transcription, indiquant un changement épigénétique coordonné régulé avec les différentes étapes du cycle cellulaires. Nous avons de plus déterminé comment le patron de méthylation du promoteur de l'ARNr est régulé

Small G proteins of the Ras superfamily were reported as major regulators of various cellular functions and are known to play a role in numerous human pathologies, such as cancer. They are currently classified according to their amino acid sequence into five subfamilies (Ras, Rho, Rab, Arf and Ran). In order to systematically study their functions and redundancy, we have selected C. elegans as an experimental model. This organism was chosen since it provides a reliable simple model for the study of small G proteins: In addition to providing experimental advantages over other higher eukaryotes, C. elegans expresses well conserved small G proteins, in limited number. We used this model to perform an integrative analysis of the biochemical activities of small G proteins and to study the role of these enzymes in the maintenance of the endoplasmic reticulum homeostasis. We have first constructed a library of vectors containing 53 C. elegans GTPase genes cloned in various expression vectors. This strategy appeared very efficient since the success rate of our cloning approach was above 89%. Using two high throughput technologies, we have then developed assays to measure enzymatic activities (GTP affinity, exchange and hydrolysis activities) of small G proteins. Integration and analysis of these enzymatic results revealed that the amino acid sequence clustering only partially represents the enzymatic characteristics of small G proteins. This suggests that a biochemical classification would be more representative of small G protein function similarities than their current amino acid sequence-based classification. To develop on this idea, we have generated and collected data related to RNAi phenotypes, cellular and tissue localizations and mRNA co-expression to integrate various functional properties and proposed a more complete clustering of the Rho subfamily of proteins. Another biochemical characteristic studied in this project was the role of small G prote / Les petites protéines G de la famille des Ras sont impliquées dans de nombreux processus biologiques et elles sont impliquées dans plusieurs pathologies humaines telles que les cancers. Elles sont présentement classifiées à partir de leur séquence en acides aminés en cinq sous familles nommées Ras, Rho, Rab, Arf et Ran. Dans le but d'étudier leurs propriétés biochimiques, leurs fonctions et leur niveau de redondance de manière systématique, nous avons choisi le nématode C. elegans comme modèle expérimental. En plus des avantages expérimentaux que confère cet organisme, les petites protéines G y sont grandement conservées et beaucoup moins abondantes que chez les mammifères. Par conséquent, nous avons utilisé cet organisme pour étudier diverses fonctions de ces protéines, proposé une nouvelle classification fonctionnelle de la superfamille des Ras et étudier le rôle des GTPases dans la régulation des voies du stress du réticulum endoplasmique. Pour réaliser ce projet, nous avons tout d'abord identifié les gènes codant pour des GTPases chez C. elegans et construit une bibliothèque de vecteurs d'expression pour 53 d'entre eux. Cette stratégie s'est avérée très efficace, avec un rendement de clonage de plus de 89%. Ensuite, nous avons développé de puissants essais enzymatiques pour mesurer leurs activités biochimiques (affinité et hydrolyse du GTP et activité d'échange) en utilisant deux technologies à haut débit. De plus, l'intégration et l'analyse de ces valeurs enzymatiques nous ont permis de démontrer que la classification usuelle des petites protéines G ne corrèle pas totalement avec l'activité enzymatique de ces dernières. Par conséquent, nous avons collecté des données relatives à la localisation tissulaire et cellulaire, aux phénotypes ARNi ainsi qu'à la co-expression des ARNm de ces gènes afin de proposer une classification fonctionnelle des petites protéines G de la sous famille des Rhos

This study used the yeast two-hybrid system to identify novel proteins interacting with the orphan receptor RVR. Two gene products were pulled out of two independent screens. We named them L5B and RVRin62. The RVRint62 was characterized in this research. RVRint62 is a partial cDNA clone of approximately 2.2 kb, which is expressed in most tissues. It interacts with both members of the RVR/Rev-erb$ alpha$ subfamily of the nuclear receptor superfamily. The domains of RVR responsible for the interaction with RVRint62 were characterized as the C-terminal of the LBD and the signature motif. The interaction domain of RVRint62 is situated in the first 300 amino acids of its N-terminus. The motif LXXLL, found in many co-regulatory proteins that interact with nuclear receptors, is also found in RVRint62. However, the LXXLL motif is not required for interaction with RVR. The potential modulating role of RVRint62 on RVR activity remains to be determined.

The products of the A genes of bacteriophages Mu of Escherichia coli and D3112 of Pseudomonas aeruginosa are required for the transposition of the double-stranded DNA genomes of their respective phages. Mutant proteins of the multifunctional Mu transposase were constructed in vitro in order to locate regions of the enzyme that may be important for function. Deletions in the Mu A gene yielded two mutant proteins that help to define the end-specific DNA-binding domain. The insertion of two or four amino acids at eight different sites produced six mutant transposase proteins that were inhibited for $ mu$Mu transposition in vivo, but not DNA binding in vitro, at 37$ sp circ$C. In addition, site-specific mutagenesis was performed to alter tyrosine 414, an amino acid which is situated in a region that displays amino acid homology to the active sites of other nicking-closing enzymes. / The A and B genes of transposable phage D3112 were cloned and sequenced in order to compare their gene products to those from the analogous genes of ccliphage Mu. (Abstract shortened by UMI.)

Negative regulation of antigen and cytokine-induced intracellular signalling pathways involves several mechanisms including anergy, receptor internalization, recruitment of protein phosphatases, etc. In certain cases, these inhibitory signals specifically target the Src and Jak families of protein tyrosine kinases. Recent work in the laboratory has shown that protein tyrosine kinase Csk and protein tyrosine phosphatase PEP form a complex that inactivates Src kinases in the context of antigen receptor signalling. In this manuscript, I demonstrate that this interaction depends on the SH3 domain of Csk and a typical SH3 domain-binding motif (PXXPXR) present within PEP. Moreover, I show that the Csk SH3 domain recognizes sequences flanking the proline-rich core for in vitro and in vivo binding. In a very different mechanism, negative regulation of Jaks seems to depend, at least in part, on a recently cloned SH2 domain-containing adapter protein, termed suppressor of cytokine signalling-1 (SOCS-1). Recent evidence suggests that the inhibitory effect of SOCS-1 on Jak catalytic activity is tightly regulated by transcriptional and post-translational mechanisms. In this report, I propose an additional regulatory mechanism in the form of translational control. I show that SOCS-1 protein levels are strongly repressed by its GC-rich 5' untranslated region. The implications of these results will be discussed.

Cruciforms are an alternative secondary structure which may be adopted by DNA containing inverted repeats, under conditions of adequate torsional strain. Inverted repeats are distributed, in a non-random fashion, throughout the genomes of prokaryotes and eucaryotes. Mounting evidence suggests that they are involved in the initiation of DNA replication. A structure-specific cruciform DNA binding protein (CBP) has previously been enriched from HeLa cells, and demonstrated to be a member of the 14-3-3 family of proteins. This thesis reports the dimethylsulfate (DMS) protection footprinting of this protein on a stable cruciform, with the goal of testing a model proposed for this interaction. The footprint obtained was not clear or reproducible enough to allow this verification, however, it does support previously identified regions of binding on the cruciform DNA. Possible explanations for the nature of the footprint obtained, and suggestions which may allow the achievement of verification of the model, are discussed.

Ectopic gene targeting is an alternative outcome of the gene targeting process in which the targeting vector acquires sequences from the genomic target but proceeds to integrate elsewhere in the genome. To explore this mechanism of recombination an assay was developed, designated as Recombination Access Mapping (RAM), that uses ectopic gene targeting coupled with two color FISH analysis. Using RAM, the integration sites of the gene targeting vector with respect to the target locus were determined for ectopic gene targeting in a murine fibroblast cell line (LTA). The distribution of integration sites was found to be bimodal during ectopic gene targeting, being either within 2--3 Mb of the target or on distinct chromosomes from the chromosome carrying the target locus. Both inter and intrachromosomal sites appeared to be equally accessible to the targeting vector with site specific variations. In contrast, illegitimate integration of a transgene was found to occur randomly in the LTA cell line. Interestingly, interphase analysis indicated that vector sequences which had integrated ectopically in other chromosomes than the target co-localized with the target locus at a significant frequency as compared to random unlinked loci. Mechanistically, the homologous recombination event that produces the extrachromosomal recombinant DNA molecule, following ectopic gene targeting or interplasmid recombination, was demonstrated to potentiate or commit the recombinant DNA to integration in the genome at the site of a double-strand break (DSB). This data suggests that extrachromosomal and chromosomal DSB repair are coupled, and therefore the integration site of the transgene during ectopic gene targeting is most likely a DSB in close proximity spatially to the target locus. The conclusion of this thesis is that RAM can be used to determine which chromosomal domains within the genome are accessible to a given genetic locus. Thus, RAM presents a new paradigm for the analysis of DN

Plasmodium falciparum is the most important causative agent of human malaria. The present study focuses on the identification and characterization of pfgcn20, a novel-member of the ATP-binding cassette protein gene family in P. falciparum. In a PCR-based cloning strategy, I employed redundant oligonucleotides directed against sequences encoding the highly conserved Walker motifs within the ATP-binding cassette, and the resulting amplified DNA fragments were used for library screening and gene walking experiments that resulted in the molecular isolation of the pfgcn20 gem. The pfgcn20 gene spans 2448 nucleotides and was predicted to encode a 95.5 kilo-daltons hydrophilic protein, PFGCN20, with two conserved ATP-binding sites but no predicted membrane spanning domains. Pfgcn20 a localized on chromosome 11 and was shown to be expressed as a 3.3 kilobase mRNA transcript. The predicted PFGCN20 protein shows significant sequence homology to a subgroup of ATP-binding canette proteins related to the Saccharomyces cerevisiae translational regulator Gcn20p, to yeast and fungi EF-3 elongation factors, and to a subfamily of ATP-binding subunits of bacterial ATP-binding cassette transporters. However, no clear function for PFGCN20 could be inferred from these homology searches. / A PFGCN20 specific polyclonal antibody was generated in rabbits and used to reveal that PFGCN20 is expressed as a 95 kilo-dalton protein throughout the entire cycle of the erythrocytic stage of P. falciparum. Localization studies showed that during the ring stage, PFGCN20 is present mainly in the parasite cell either in the cytoplasm and/or in discontinuous electron dense compartments at the cell periphery. In contrast, during the trophozoite and schizont stages, PFGCN20 was found to be transported into the host cell cytoplasm where it was associated with membranes of the tubovesicular membrane system and the erythrocytic plasma membrane. In addition, a fraction of PFGCN20 appeared in the red cell cytoplasm in a membrane-free form. These results suggest that at least part of PFGCN20 function is related to physiological processes in the cytoplasm of infected red cells. / In the final chapter of my thesis, I have analyzed the functional homology between PFGCN20 and Gcn20p, by expression of PFGCN20 in the B2511 S. cerevisiae GCN20 deletion strain. Similar to Gcn20p, overproduction of PFGCN20 in H2511 (Deltagcn20) partially restored yeast cell tolerance to, an inhibitor of histidine biosynthesis, 3-amino-1,2,4-triazole. Expression of truncated pfgcn20 and chimeric pfgcn20/GCN20 constructs revealed that the N-terminal segment (211 acids) of PFGCN20 is sufficient for the restoration of partial 3-amino-.1,2,4-triazole tolerance of H2511 (Deltagcn20 ) cells. From my combined results, I have derived the working hypothesis that PFGCN20 may function as translational regulator within the plasmodial cell and fulfills a second possibly transport-related, function in the cytoplasm of infected red blood cells. However, whether PFGCN20 truly represents an example of a bi-functional ATP-binding protein must be evaluated by additional experimental work.

The tumor suppressor p53 is a multifunctional protein that plays a critical role in modulating cellular responses such as cell cycle arrest and apoptosis upon DNA damage or other stresses. These functions of p53 can be regulated by protein-protein interactions and by phosphorylation. The interferon (IFN)-inducible, double-stranded RNA activated protein kinase PKR is a serine/threonine kinase that mediates the antiviral and antigrowth effects of IFNs by inhibiting protein synthesis through the phosphorylation of the translation initiation factor eIF-2alpha. Both PKR and p53 have the capacity to inhibit cell growth and to modulate gene expression at both the transcriptional and translational level. However, the exact molecular mechanism(s) behind the regulation of cell growth by PKR are not-well understood. Based on previous reports demonstrating a cross talk between other IFN-inducible proteins and p53, it was of interest to examine whether and how PKR might modulate p53 function to explain the ability of PKR to regulate cell growth. This study demonstrates that PKR can physically, functionally and genetically interact with p53. PKR and p53 physically interact both in vivo and in vitro, and the interaction of PKR with p53 is a functional one since PKR can directly phosphorylate p53 on serine392 in vitro. Furthermore, studies using PKR+/+ and PKR-/- cells treated with adriamycin or expressing a temperature sensitive mutant of p53, showed that p53-mediated gene expression and G0/G 1 arrest was inhibited in PKR-/- cells. The phosphorylation levels of serine18 of p53, which plays a role in p53 function in vivo, was inhibited in PKR+/+ cells expressing temperature sensitive p53 compared to PKR-/- cells. Induction of serine18 phosphorylation was inhibited in PKR-/- cells in response to gamma-radiation and adriamycin but not ultraviolet radiation. While PKR does not directly phosphorylate serine18 of p53 in vitro, the specific phosphatidylinositol-3 (PI-3) kinase inhibit

The study of how antigens and superantigens are presented to T cell has always been a great challenge to immunologists. Mainly because the understanding of how subtle changes in the interaction of the T cell receptor (TCR) with its ligands affect the development of an immune response opens the possibility that this knowledge could be used to treat various immunopathological conditions. Bacterial superantigens are of clinical relevance since they cause diseases in human, but also for their potential as immunotherapeutic agents. More fundamentally, SAgs have always been valuable tools in gaining insights into the mechanisms of T cell activation by major histocompatibility complex (MHC)/peptide ligands. The work presented in this thesis addresses structural, as well as functional aspects of the interaction between SAg, major histocompatibility class II molecules and T cell receptor. Finally, some of the work presented made use of SAgs as tools to investigate the precise mechanics by which the physiological engagement of the TCR is interpreted into a qualitatively diverse functional response from T cells.