Eukaryotic cells repair DNA double-strand breaks (DSBs) through both non-homologous and homologous recombination pathways. The initiation of homologous recombination requires the generation of 3' overhangs, which are essential for the formation of Rad51 protein-DNA filaments that catalyze subsequent steps of strand invasion. Experiments in budding yeast show that resection of the 5' strand at a DSB is delayed in strains lacking any components of the Mre11/Rad50/Xrs2 (MRX) complex¹ . In meiosis, a specific class of hypomorphic mutants of mre11 and rad50 (Rad50S) are completely deficient in 5' resection and leave Spo11 covalently attached to the 5' strands of DNA breaks². Similar to mre11S and rad50S mutants, sae2 deletion strains fail to resect 5' strands at meiotic DSBs and accumulate covalent Spo11 adducts³;⁴. In addition, Sae2 and MRX were also found to function cooperatively to process hairpin-capped DNA ends in vivo in yeast. sae2 and mrx null strains show a severe defect in processing these structures and accumulate hairpin-capped DNA ends⁵;⁶. The Longhese laboratory has also shown that Sae2 deletion strains show a delay in 5' strand resection, similar to rad50S strains⁷. Recently, Bettina Lengsfeld in our laboratory demonstrated that Sae2 itself possesses nuclease activity and that MRX and Sae2 act cooperatively to cleave single-stranded DNA adjacent to DNA hairpin structures⁸. In vitro characterization of Sae2 showed that the central and N-terminal domains are required for MRX-independent nuclease activity and that the C-terminus is required for cooperative activities with MRX. Sae2 also acts independently of MRX as a 5' flap endonuclease on branched structures in vitro. Our studies investigate whether MRX, Sae2, and Exo1 function cooperatively in DNA resection using recombinant, purified proteins in vitro. We developed assays utilizing strand-specific Southern blot analysis to visualize DNA end processing of model DNA substrates using recombinant proteins in vitro. Our results demonstrate that MRX and Sae2 cooperatively resect the 5' end of a DNA duplex together with the Exo1 enzyme, supporting a role for these factors in the early stages of homologous recombination and repair. / text

Les télomères sont des structures nucléoprotéiques spécialisées dont l'un des rôles est d'empêcher le raccourcissement progressif de l'extrémité des chromosomes suite à la réplication et à l'instabilité génomique due à la recombinaison de l'extrémité de chromosomes. Malgré le rôle des télomères dans la protection de l'extrémité des chromosomes contre les mécanismes de réparation de l'ADN et de recombinaison, de nombreuses protéines de ces voies jouent des rôles essentiels dans l'homéostasie télomérique et la stabilité des chromosomes. Parmi elles, la protéine RAD50 appartenant au complexe MRE11/RAD50/XRS25(NBS1) et l'endonucléase structure spécifique XPF/ERCC1 sont localisées aux télomères ; ces deux complexes connus pour leur rôle dans les voies de réparation de l'ADN ainsi que dans les études sur la recombinaison. Nous avons identifié deux rôles différents pour la protéine RAD50 dans la maintenance télomérique et dans la protection des extrémités des chromosomes, en contexte de présence et absence de la télomérase. L'absence d'AtRAD50 augmente significativement le nombre de fusions chromosomiques impliquant des télomères raccourcis. Nous proposons que ce rôle protecteur des télomères raccourcis de RAD50 est le résultat de sa fonction de contraindre la recombinaison entre chromatides soeurs et ainsi d'éviter les évènements de fusions par les extrémités. Nous avons recherché le ou des mécanismes impliqué(s) dans ces évènements de fusions chromosomiques chez les mutants atrad50 en réalisant des croisements entre les plantes déficientes pour ATRAD50 et des plantes déficientes pour des gènes codant des protéines des voies de réparation par recombinaison non-homologue et homologue. Au contraire de la situation en cellules de mammifères, nous n'avons pas observé d'instabilité chromosomique chez les plantes mutantes correspondantes pour XPF (AtRAD1) or ERCC1 (AtERCC1). Cependant, en absence de la télomérase, la mutation de l'un de ces deux gènes entraîne une augmentation précose et significative de l'instabilité chromosomique sans accélération générale de la perte des répétitions télomériques, mais associée à la présence de fragments ADN extrachromatiques visibles en cytologie. Une analyse intensive par FISH a permis d'identifier ces ADN comme des bras entiers spécifiques de deux chromosomes. Nos données indiquent un rôle protecteur de RAD1/ERCC1 comme l'invasion de l'ADN simple brin télométrique dans des séquences télomériques interstitielles. Le fait que les mutations de rad1 (ou ercc1) augmentent dramatiquement l'instabilité chromosomique des mutants télomérase a des implications très importantes pour les modèles des rôles de la recombinaison aux télomères.

Meiotic recombination is a prerequisite for exchange of genetic information in all Sexually reproducing organisms. This process is initiated by the formation of double stranded breaks (DSBs) in DNA followed by homology directed repair. The process is subjected to surveillance mechanisms that control DSB formation and allow for repair of DSBs by halting cell cycle progression. Interestingly, though generation of DSBs is an Essential event in meiosis they are nevertheless regarded as the most lethal forms of DNA damage. If left unrepaired a single DSB can lead to gene deletion, duplication, translocations and missegregation of large chromosome fragments leading to cell death. In Saccharomyces cerevisiae, genetic screens for mutants defective in meiotic recombination led to the identification of a group of genes called the RAD52 epistasis group which includes RAD50, RAD51, RAD52, RAD54, RAD55, RAD57, RAD59, MRE11 and XRS2. A subset of these genes, namely MRE11, RAD50 and XRS2, have been shown by genetic studies to be essential for several nuclear events including sensing DSBs, double strand break repair (DSBR) by homologous recombination (HR) and non homologous end joining (NHEJ), telomere length maintenance, cell cycle activation in response to DSBs, mitotic and meiotic recombination. In vitro, Mre11 displays Mn2+-dependent endonuclease activity on ssDNA, 3'-5' Exonuclease on single- and double-stranded DNA, strand annealing and weak hairpin Opening activities. Mutational analyses have revealed two functional domains in Mre11- Then terminal nuclease domain involved in telomere length maintenance and DSB Processing and the C terminal DNA binding domain involved in DSB formation during Meiosis. Rad50, a 153 kDa protein shares homology with the SMC (Structural Maintenance of Chromosome) family of proteins which are involved in chromosome Condensation and cohesion. It consists of a bipartite N- and C terminal Walker A and Walker B motifs separated by a heptad repeat sequence which folds into an antiparallel Coiled-coil structure. The heptad repeats are separated by a metal binding globular region the Zn hook. Rad50 is an ATP-dependent DNA-binding protein. hRad50 regulates the exonuclease activity of hMre11. Unlike Mre11 and Rad50, which are evolutionarily conserved, Xrs2 is found only in S. cerevisiae and Nbs1 in mammals. Xrs2 appears to be sequence non-specific DNA- binding protein. Xrs2 in yeast or Nbs1 is its counterpart in mammals target Mre11 and Rad50 to the sites of DNA damage and mediate S-phase cell cycle checkpoint activation. Mutations in either one of the MRX subunits results in defects in repair of DSBs, activation of cell cycle checkpoint and shortened telomeres leading to genomic instability. Hypomorphic mutations in MRE11 and NBS1 lead to genetic disorders- A-TLD (ataxia-telangiectasia-like disorder) and NBS (Nijmegen breakage syndrome) respectively, that are phenotypic ally related to AT (ataxia-telangiectasia) caused by mutations in ATM. Patients with AT, A-TLD or NBS syndromes are hypersensitive to radiomimetic agents and are predisposed to cancer. Several lines of evidence suggest that S. cerevisiae strains bearing mre11Δ, rad50Δ or xrs2Δ display shortening of telomeres. Telomeres are the nucleoprotein ends of all linear eukaryotic chromosomes that are important in maintaining the integrity of the genome.Telomeres are comprised of repetitive G rich sequence most of which is double stranded but the extreme 3' end protrudes to form 3' single stranded overhang called the G tail. elopers are essential in preventing end-end fusion of chromosome, are important for chromosome replication, segregation and genome stability. Genetic studies have implicated the MRX complex in both telomerase-dependent and independent telomere length maintenance. Studies have indicated a direct role for S. cerevisiae MRE11 in the proper establishment of telomere end-structure. However, the molecular mechanism of MRX at telomeres is poorly understood. To understand the role(s) of MRX complex at telomeres, it is important to elucidate the biochemical activities of MRX complex as well as its individual subunits on the telomere DNA structures. Since, Mre11 complex is known to function in several processes related to DNA metabolism it becomes imperative to study the function of Mre11 complex on DNA substrates in the context of a given nuclear process. The 3' single trended telomeric sequence is capable of acquiring folded conformation(s) as a mechanism of end protection which is mediated by several telomere-specific and nonspecific ending proteins. In mammals, the 3' ssDNA has been demonstrated to fold into tloop configuration mediated by some of the components of sheltrin protein complex, wherein the ssDNA invades the duplex DNA resulting in the formation of a displacement loop (D loop). Evidence for the formation of t-loop has been shown in vitro with human telomeres. However, the formation of t-loops has not been demonstrated in S. cerevisiae. Nevertheless, there is growing body of evidence which suggests the formation of alternative DNA structures such as G4 DNA at the yeast telomeres. G quadruplexes (G quartets or G4 DNA) are thermodynamically stable structures formed by Hoogsteen base pairing between guanine residues. In a G quartet the four guanine residues are paired, where each guanine residue is an electron acceptor and a donor and stabilized by a metal cation. The presence of G rich motifs at the promoter regions, rDNA, telomeres and recombination hot spots indicate that G4 DNA has important functions in vivo. Although the existence of G4 DNA has been the subject of much debate, the identification of several proteins that promote (Rap1, Hop1, Topo I, TEBPβ), modify and resolve (POT1, TERT, KEM1, GQN1, BLM, WRN, Rte1) G4 DNA, together with the direct visualization of G4 DNA using G4 DNA specific antibodies and RNA interference have provided compelling for the existence of G4 DNA in vivo. To elucidate the function of MRX complex or its individual subunits at telomeres, the biochemical activities of purified MRX complex and its individual subunits on G4 DNA, D loop, duplex DNA and G rich ssDNA has been analyzed in this study. G4 DNA was assembled from S. cerevisiae telomeric sequence. G4 DNA was isolated and its identity was ascertained by chemical probing and circular dichroism. S. cerevisiae MRE11 and XRS2 was cloned and expressed in E. coli BL21 (DE3)plysS. S. cerevisiae RAD50 in pPM231 vector in S. cerevisiae BJ5464 strain was a gift from Dr. Patrick Sung (Yale University). Mre11, Rad50 and Xrs2 were overexpressed and purified to >98% homogeneity. The identity of the proteins was ascertained by Western bloting using polyclonal antibodies. Using purified proteins heterotrimeric MRX and heterodimeric MR and MX protein complexes were formed in the absence of ATP, DNA or Mn2+. The ability of M/R/X to bind to telomeric DNA substrates was studied by electrophoretic mobility shift assays. Mre11, Rad50, Xrs2 and MRX displayed higher binding affinity for G4 DNA over D loop, ss- or dsDNA. MRX bound G4 DNA more efficiently compared to its individual subunits as 10-fold lower concentration of MRX was able to shift the DNA into the protein-DNA complex. The protein-G4 DNA complexes were stable as >0.8 M NaCl as required to dissociate 50% of protein-G4 DNA complexes. Efficient competition by poly(dG), which is known to fold into G4 DNA, suggested that the protein-G4 DNA complex was specific. Competition experiments with tetra-[N-methyl- pyridyl]-porphyrin suggested that M/R/X recognizes distinct determinants and makes specific interactions with G4 DNA. G4 DNA is highly polymorphic and can exist as intramolecular or intermolecular (parallel and antiparallel) structures. High affinity binding of Mre11 to G4 DNA (parallel) over G2' DNA (antiparallel), ss- and dsDNA suggests the existence of parallel G4 DNA structures at the telomeres and that G4 DNA may be the natural substrate for MRX complex in vivo. Telomeres are elongated by telomerase that requires access to the 3' G-tail for its activity. Formation of G4 DNA structures renders the 3' G-tail inaccessible to telomerase thereby inhibiting telomere elongation. To elucidate the functional relevance of high affinity of M/R/X for G4 DNA, the ability of the complex to generate the appropriate DNA structure for telomere elongation has been analyzed. In this study, I considered the possibility that MRX could act as: (a) a helicase that opens up the G4 DNA structures making it accessible to telomerase or (b) as a nuclease that cleaves the G4 DNA generating substrates for telomerase. Helicase assay with Mre11, Xrs2, MX and MRX on G4 DNA and duplex DNA showed no detectable DNA unwinding activity. Interestingly, nuclease assays with Mre11 on G4 DNA showed that Mre11 cleaved G4 DNA in Mn2+-dependent manner and the cleavage was mapped to the G residues at the stacks of G quartets. Mre11 cleaved telomeric duplex DNA in the center of TGTG repeat sequence, G rich ssDNA at 5' G residue in an array of 3 G residues and D loop structure preferentially at the 5' ends at TG residues. Significantly, the endonuclease activity of Mre11 was abrogated by Rad50. Xrs2 had no effect on the endonuclease activity of Mre11. Structural studies on Rad50 and Mre11 showed that binding of ATP by Rad50 positions the Rad50 catalytic domain in close proximity to the nuclease active site of Mre11. In yeast, disruption of ATP binding Walker motifs results in a null phenotype, suggesting that ATP is required for Rad50 functions in vivo. hRad50 is known to regulate the exonuclease activity of hMre11 in the presence of ATP. Therefore, can ATP modulate the effect of S. cerevisiae (Sc) Rad50 on ScMre11? To address this question, I monitored the ATPase activity of Rad50 in the absence or presence of DNA. Rad50 hydrolyzed ATP in a DNA-independent manner; however, ATPase activity was enhanced in the presence of Mre11 and Xrs2. However, Rad50 exhibited a low turnover indicating that ATP could function as a switch molecule. Based on these observations, the effect of ATP on the nuclease activity was examined. The binding of ATP and its hydrolysis by Rad50 attenuated the inhibition exerted by Rad50 on the Mre11 endonuclease activity. Cleavage of G4 DNA, D loop, duplex DNA and ssDNA required ATP hydrolysis, since no cleavage product was observed when ADP or ATPγS was substituted for ATP. This observation was corroborated using a hairpin DNA substrate that mimics a intermediate in VDJ recombination, thereby confirming the generality of regulation of Rad50 on the endonuclease activity of Mre11. Does Rad50 regulate the exonuclease activity of Mre11 as well? To address this question, exonuclease activity of Mre11, MR and MRX on 3' labeled duplex DNA and G4 DNA was assayed. Rad50 had no measurable effect on the exonuclease activity of Mre11. Based on previous studies and my observations, I propose a model for the role of MRX in telomere length maintenance and its regulation by the ATP-binding pocket of Rad50. MRX binds telomeric DNA substrates in a non-productive complex, which is converted to a catalytically active complex upon binding of ATP by Rad50. ATP induces conformational changes, repositioning the complex such that the catalytic site of Mre11 now has access to the substrate. Following cleavage of DNA by Mre11, the release of ADP and inorganic phosphate, generate the cleaved product. The cleaved DNA is now accessible to telomerase or telomere binding proteins. In summary, the data presented in my PhD thesis demonstrates that Mre11 is a structure- and sequence-specific endonuclease. The natural substrate for telomerase is the 3' ssDNA. G quartets at telomeres not only protect the ends from degradation but also make the ends inaccessible for telomerase activity. Genetic studies have shown that cells proficient for telomerase activity but lacking any one of the components of the MRX complex display shortening in telomere length. The ability of Mre11 to cleave G4 DNA at the stacks of G quartets therefore, suggests a mechanism by which the 3' ssDNA is rendered accessible to telomerase or other telomere binding proteins. Yeast telomeres are characterized by the presence of subtelomeric Y' elements proximal to the terminal TG1- 3 repeat sequences. The Y' element has been shown to be amplified by telomerase in a fraction of mutants with short telomeres. The mechanism by which Y' DNA is amplified is unclear. The ability of Mre11 to cleave telomere duplex DNA at the center of TGTG repeats could contribute to the generation of appropriate substrate for elongation by telomerase, thereby contributing to Y' DNA amplification. Telomere length is maintained by homeostasis between processes that contribute to telomere elongation and those that cause attrition in telomeric ends. Overelongated telomeres are brought to wild type telomere size by a unique recombinational single step deletion process termed telomere rapid deletion (TRD). TRD involves invasion of the elongated 3' G tail into the proximal telomeric tract resulting in the formation of the D loop structure. Following branch migration the D-loop is nicked and resolved into a deleted telomere and a circular liner product. Cells deleted for MRE11, RAD50 or XRS2 are deficient in TRD process. It has been hypothesized that Mre11 could be a candidate for cleaving the D-loop structure. The endonuclease activity of Mre11 on D-loop structure, preferentially at the 5' ends at TG residues demonstrated in this study, show that Mre11 could function as the nuclease required to generate the deleted telomere in TRD. MRX complex is involved in several processes involving DNA metabolism. It is important that the activities of the complex are regulated in the in vivo context. Complex formation and the interaction of the individual subunits with nucleotide cofactors and metal ions constitute a mode of regulation. This study shows that Rad50 regulates the endonuclease, but not exonuclease activity of Mre11. The binding of ATP and its hydrolysis by Rad50 brings in the regulatory factor necessary to keep the uncontrolled nuclease activity of MRX in check, thus preventing any deleterious effects on telomere length. Telomere maintenance by telomerase is activated in 80% of cancer cells. Inhibition of telomerase by G quartets provides a new drug targets for potential anti-cancer drugs. It is, therefore, likely that understanding the biological consequences of G quadruplex interactions would provide a better insight in development of therapeutics for cancer.

L'ADN de chaque cellule est constamment soumis à des stress pouvant compromettre son intégrité. Les bris double-brins sont probablement les dommages les plus nocifs pour la cellule et peuvent être des sources de réarrangements chromosomiques majeurs et mener au cancer s’ils sont mal réparés. La recombinaison homologue et la jonction d’extrémités non-homologues (JENH) sont deux voies fondamentalement différentes utilisées pour réparer ce type de dommage. Or, les mécanismes régulant le choix entre ces deux voies pour la réparation des bris double-brins demeurent nébuleux. Le complexe Mre11-Rad50-Xrs2 (MRX) est le premier acteur à être recruté à ce type de bris où il contribue à la réparation par recombinaison homologue ou JENH. À l’intersection de ces deux voies, il est donc idéalement placé pour orienter le choix de réparation. Ce mémoire met en lumière deux systèmes distincts de phosphorylation du complexe MRX régulant spécifiquement le JENH. L’un dépend de la progression du cycle cellulaire et inhibe le JENH, tandis que l’autre requiert la présence de dommages à l’ADN et est nécessaire au JENH. Ensembles, nos résultats suggèrent que le complexe MRX intègre différents phospho-stimuli pour réguler le choix de la voie de réparation. / The genome of every cell is constantly subjected to stresses that could compromise its integrity. DNA double-strand breaks (DSB) are amongst the most damaging events for a cell and can lead to gross chromosomal rearrangements, cell death and cancer if improperly repaired. Homologous recombination and non-homologous end joining (NHEJ) are the main repair pathways responsible for the repair of DSBs. However, the mechanistic basis of both pathways is fundamentally different and the regulation of the choice between both for the repair of DSBs remains largely misunderstood. The Mre11-Rad50-Xrs2 (MRX) complex acts as a DSB first responder and contributes to repair by both homologous recombination and NHEJ. Being at the crossroads of both DSB repair pathways, the MRX complex is therefore in a convenient position to influence the repair choice. This thesis unravels two distinct phosphorylation systems modifying the MRX complex and specifically regulating repair by NHEJ. The first relies on cell cycle progression and inhibits NHEJ, while the second requires the presence of DNA damage and is necessary for efficient NHEJ. Together, our results suggest a model in which the MRX complex would act as an integrator of phospho-stimuli in order to regulate the DSB repair pathway choice.

L'ADN de chaque cellule est constamment soumis à des stress pouvant compromettre son intégrité. Les bris double-brins sont probablement les dommages les plus nocifs pour la cellule et peuvent être des sources de réarrangements chromosomiques majeurs et mener au cancer s’ils sont mal réparés. La recombinaison homologue et la jonction d’extrémités non-homologues (JENH) sont deux voies fondamentalement différentes utilisées pour réparer ce type de dommage. Or, les mécanismes régulant le choix entre ces deux voies pour la réparation des bris double-brins demeurent nébuleux. Le complexe Mre11-Rad50-Xrs2 (MRX) est le premier acteur à être recruté à ce type de bris où il contribue à la réparation par recombinaison homologue ou JENH. À l’intersection de ces deux voies, il est donc idéalement placé pour orienter le choix de réparation. Ce mémoire met en lumière deux systèmes distincts de phosphorylation du complexe MRX régulant spécifiquement le JENH. L’un dépend de la progression du cycle cellulaire et inhibe le JENH, tandis que l’autre requiert la présence de dommages à l’ADN et est nécessaire au JENH. Ensembles, nos résultats suggèrent que le complexe MRX intègre différents phospho-stimuli pour réguler le choix de la voie de réparation. / The genome of every cell is constantly subjected to stresses that could compromise its integrity. DNA double-strand breaks (DSB) are amongst the most damaging events for a cell and can lead to gross chromosomal rearrangements, cell death and cancer if improperly repaired. Homologous recombination and non-homologous end joining (NHEJ) are the main repair pathways responsible for the repair of DSBs. However, the mechanistic basis of both pathways is fundamentally different and the regulation of the choice between both for the repair of DSBs remains largely misunderstood. The Mre11-Rad50-Xrs2 (MRX) complex acts as a DSB first responder and contributes to repair by both homologous recombination and NHEJ. Being at the crossroads of both DSB repair pathways, the MRX complex is therefore in a convenient position to influence the repair choice. This thesis unravels two distinct phosphorylation systems modifying the MRX complex and specifically regulating repair by NHEJ. The first relies on cell cycle progression and inhibits NHEJ, while the second requires the presence of DNA damage and is necessary for efficient NHEJ. Together, our results suggest a model in which the MRX complex would act as an integrator of phospho-stimuli in order to regulate the DSB repair pathway choice.