No description available.

This thesis concentrates on understanding how individual nonspecific DNA–protein contacts are used in the excision mechanism of the human DNA repair enzyme, alkyladenine DNA glycosylase (AAG). Initially, studies focus on understanding the structure and magnitude of these fundamentally different DNA–protein stacking and T-shaped interactions to be applied to the active site of AAG. High-level ab initio techniques revealed fundamental knowledge about the structure and magnitude of these distinctly different – and +– contacts between (one or two) conjugated amino acid(s) and one nucleobase. Additionally, the mechanism used by AAG to excise (neutral and cationic) damaged nucleotides was investigated using a hybrid ONIOM approach. Reaction potential energy surfaces reveal that AAG prefers to excise both neutral and cationic substrates through a concerted mechanism, yet the nonspecific contacts present in the active site are only catalytic for the cleavage of the neutral substrates. / xvi, 195 leaves : ill. (some col.) ; 29 cm + 1 CD-ROM

En phase somatique, plusieurs mécanismes de réparations de l’ADNinterviennent pour réparer les cassures double brin (CDB) de l’ADN. Enphase méiotique, les CDB de l’ADN engendrées de façon programmées parSpo11 sont réparées par la recombinaison homologue (RH) dont les acteursprincipaux sont Rad51 et Dmc1 aidés de Brca2. Chez Arabidopsis, en absencede Brca2, le déroulement méiotique est perturbé, les chromosomes nes’associent pas en bivalents, ils apparaissent emmêlés. Ainsi, en absencede Brca2, la recombinaison homologue pourrait ne plus être fonctionnelleet les anomalies chromosomiques observées pourraient être le résultat deréparations aberrantes des CDB de l’ADN effectuée par d’autres mécanismesde réparation de l’ADN. Nous avons montrés, chez Arabidopsis, que le Nonhomologous End Joining (NHEJ) et/ou le Single Strand Annealing (SSA),mécanismes de réparation des CDB de l’ADN en phase somatique,n’intervenaient pas en phase méiotique dans la formation des anomaliesobservées en absence de Brca2. Toujours dans l’hypothèse où ces figuresméiotiques soient le résultat de liaisons covalentes, nous avons regardési les ADN-ligases ne pourraient pas être impliquées. Ainsi, nous avons pumontrer que la Ligase 6, ADN-ligase spécifique des plantes, n’avait pas derôle dans les anomalies chromosomiques observées en méiose en absence deBrca2. D’ailleurs la Ligase 6 ne semble pas non plus intervenir dans lesfigures chromosomiques observées chez les mutants rad51 et mnd1. Le rôlede la Ligase 6 n’ayant pas été déterminé lorsque nous avons démarré cetravail, nous avons voulu identifier son le rôle en étudiant le mutantcorrespondant. Le mutant ligase 6 ne présente pas de sensibilité auxstress génotoxiques utilisés ce qui indique que la Ligase 6 ne semble pasintervenir dans la réparation de l’ADN. La mutation dans le gène LIGASE Iest létal à l’état homozygote, de plus nous avons pu observer uneségrégation anormale chez l’hétérozygote mutant pour le gène LIGASE I. Lalétalité du mutant ligase I a été contournée par l’utilisation d’unsystème ARNi pour éteindre l’expression du gène LIGASE I uniquement enméiose. Cependant, l’implication de la Ligase I, dans les anomaliesméiotiques observées en absence de Brca2 n’a pas pu être déterminée.Enfin, nous avons confirmé que, chez Arabidopsis, Xrcc4 avait un rôle dansle NHEJ via son interaction avec la Ligase IV et via la sensibilité dumutant xrcc4 à différents stress génotoxiques. En revanche, Xrcc4-like nesemble pas interagir avec les acteurs du complexe de ligation du NHEJ etle mutant ne présente pas de sensibilité aux stress génotoxique, indiquantque cette protéine n’est pas impliquée dans le NHEJ et plus généralementdans les mécanismes de réparation de l’ADN. / In somatic cells, several mechanisms are involved in the repair of DNAdouble strand breaks (DSB). In meiotic cells, programmed DSBs are causedby Spo11 and repaired by homologous recombination (HR), whose main playersare Rad51 and Dmc1 aided by Brca2. In Arabidopsis, in the absence ofBrca2, meiosis is disturbed, chromosomes do not organize into bivalents,they appear stuck and entangled together. Thus, in the absence of Brca2,HR may be no functional and the chromosomal anomalies we observecouldresult from the aberrant repair of the DNA DSBs due to other mechanisms ofDNA repair. We have shown in Arabidopsis that the homologous end joining(NHEJ) and/or Single Strand Annealing (SSA), mechanisms of DNA DSB repairthat are active in the somatic phase, were not involved in the formationof the meiotic anomalies observed in the absence of Brca2 in meioticcells. Still assuming that these figures are the result of meioticcovalent bond, we checked whether DNA ligases could be involved. Thus, wehave shown that 6 Ligase, DNA ligase specific plants, had no role in thechromosomal abnormalities observed in meiosis in the absence of Brca2.Besides the Ligase 6 does not seem to interfere with the meiotic figuresobserved in rad51 and mnd1 mutants. We wanted to identify the Ligase 6role in studying its mutant. Ligase 6 mutant did not show sensitivity togenotoxic stress. The Ligase 6 does not seem to be involved in DNA repair.The lethality of the ligase I mutant was bypassed with a RNAi constructaimed at extinguishing the gene expression of LIGASE I atmeiosis only.However, the involvement of Ligase I in the meiotic anomalies observed inthe absence of Brca2 could not be determined. Finally, we confirmed that,in Arabidopsis, Xrcc4 has a role in NHEJ through its interaction withligase IV and the sensitivity of the xrcc4 mutant to different genotoxicstress. In contrast, Xrcc4-like does not appear to interact with playersin the NHEJ ligation complex and the mutant shows no sensitivity togenotoxic stress. These result indicated that this protein is not involvedin NHEJ and, more generally in the mechanisms of DNA repair.

Repairing DNA damage is brought about by highly specific proteins that partake in a variety of DNA repair. Two of the most common types of damage are double-strand breaks (DSBs) and interstrand crosslinks. A single DSB or crosslink can potentially kill a cell if it is not repaired~ In human and other vertebrate cells, DSBs are repaired by two different mechanisms. The nonhomologous end-joining pathway can bring together the broken ends and join them, usually with the loss of some nucleotide sequence. A second pathway, homologous recombinational repair (HRR), is equally important. This repair process utilizes the information provided by another DNA molecule to restore damaged DNA. This molecule is usually a sister chromatid arising from DNA replication. This process is essentially error-free, unlike the end-joining process. Some HRR activity is required for proliferating cells to remain viable. The central protein player is RAD51, which with the help of other proteins such as XRCC2, XRCC3, RAD51B, RAD51C, and RAD51D, performs the critical initiating steps of homologous pairing and strand transfer. The proteins encoded by the familial breast cancer genes, brcal and brca2, also play an important role in HRR. My project is concerned with studying proteins that interact with Drosophila melanogaster (XRCC2). Proteins interacting with DmXRCC2 were identified by using a yeast two hybrid system. "Bait fusion protein" (DmXRCC2 linked to GAIA BD) was constructed by Dr. Wrischnik. Tanya Dimetrijevich, a graduate student, used this bait to fish for interacting or "target" proteins. About 50 such proteins were found. I began validating these target proteins with the intention of exploring novel interactions and functions of DmXRCC2. The process of validating proteins interacting with DmXRCC2 yielded two very interesting candidate proteins-CaBPl and FAF. · CaBPl, also called protein disulfide isomerase P5, is an endoplasmic-reticulum calciumbinding protein. FAF belongs to a large family of deubiquitinating enzymes that cleave ubiquitin-protein bonds and play diverse roles in the ubiquitin pathway. One of the implications of such discoveries could be to compare and contrast DmXRCC2 and human XRCC2 in terms of their interactions and functions.

Genomic integrity is important for living cells' correct functioning and propagation. Deoxyribonucleic acid as a molecule is a subject to chemical reactions with agents that can come from environment as well as from internal metabolism processes. These reactions can induce damage to DNA and thus compromise the genetic information, and result in disease and death of an organism. To mitigate the damage to DNA, cells have evolved to have multiple DNA repair pathways. Presented here is a computational study of DNA repair genes. The structure of the Homo sapiens direct DNA repair gene ALKBH1 is predicted utilizing homology modeling methods and using AlkB and DBL proteins as templates. Analysis of the obtained structure and molecular dynamics simulations give insights into potentially functionally important residues of the protein. In particular, zinc finger domains are predicted, and lysines that could perform catalytic activities are investigated. Subsequent mutagenesis experiments revealed the effect of the residues predicted to form zinc fingers on activity of ALKBH1. Structure and dynamics of AlkD, a Bascillus cereus base excision DNA repair protein is also studied. This protein has been shown to bind DNA with large alkyl adducts and perform excision catalysis without base flipping which is characteristic to other enzymes in the same family. MD simulations of AlkD revealed that B helix, which interacts with DNA, has higher fluctuations when AlkD is not bound to DNA, and thus could have a role in binding and recognition of DNA. For the purpose of finding biomarkers and to further our understanding of a mode of action of DNA repair genes, statistical methods were applied to identify mutations that are linked to cancer phenotypes. Analysis was based on case-control studies of patients with cancers of prostate, breast, pancreas, lung as well as chronic lymphocytic leukemia from NCBI dbGAP database. Those mutations that result in missense mutations were further investigated. In particular, extensive MD simulations and experimental investigations were performed on the mutation in the ALKBH7 gene that was found to be linked to prostate cancer.