No description available.

Summary in Chinese. / Thesis (M.S.)--Chinese University of Hong Kong. / Bibliography: l. 84-95.

Telomeres are the nucleoprotein structures that protect the ends of eukaryotic chromosomes from being recognized as DNA double-strand breaks. Telomeres are recognized by the ribonucleoprotein telomerase, a reverse transcriptase that catalyzes addition of G-rich telomeric DNA repeats to the 3’ overhang of the telomere. The action of telomerase allows cells to overcome the end-replication problem defined by the inability of conventional DNA polymerases to fully replicate the end of the chromosome. Telomeric DNA tracts are maintained in a species-specific size range primarily through the competition between telomerase and the end-replication problem. In many organisms, recombinational activities can function at telomeres outside of the wild type range, in some cases resulting in telomerase-independent telomere maintenance. Telomere rapid deletion (TRD) can dramatically shorten elongated telomeres. Elongation of telomeres below the normal range in the absence of telomerase is known as alternative lengthening of telomeres (ALT). Here we demonstrate that telomeres in Arabidopsis thaliana are also subjected to these recombinational activities. Elongated telomeres in ku70 mutants are shortened by TRD. In contrast to other organisms, TRD functions on telomeres of wild type length. TRD produces extra-chromosomal telomeric circles, which can serve as substrates for ALT. In Arabidopsis, ALT may require the byproducts of TRD, as telomerase mutants with extremely short telomeres are unable to maintain telomeric repeats by recombination and instead secure their genome through an unknown mechanism. Finally, we follow the fate of cells with telomere-to-telomere fusions. Fusions are not propagated to viable progeny. We propose that a G1 checkpoint dependent upon the checkpoint protein ATM arrests cells following the break of a single telomere fusion. We design reporter constructs to follow the fate of individual cells with telomere fusions, and present initial characterization of their expression. We find no evidence for the propagation of telomere fusions in somatic cells, though later generation mutants will provide a better test of this hypothesis. This work begins the study of the fate of cells with telomere fusions in Arabidopsis. Furthermore, it sets the foundation for studying recombinational shortening and elongation of telomeres in Arabidopsis and the effects of these processes on telomere length regulation.

Dysgenic male recombination (MR) induced by the P strains T-007 and OKI rarely, if ever, occurs in the heterochromatin of chromosome two. One possible explanation is that the lack of heterochromatic exchange is due to the highly condensed chromatin in this region. Butyrate (a suspected modifier of chromatin structure) induced significant levels of heterochromatic MR in dysgenic hybrids derived from crosses involving two different P strains. This finding is consistent with the hypothesis that chromatin structure can influence the insertion and excision of P elements and hence MR. Analogous experiments were performed using third chromosome suppressor of variegation (Su(var)) mutations. Neither suppressor mutation induced any heterochromatic MR, suggesting that the mode of action of these Su(var) genes is different from, and more specific than, that of butyrate. One of the mutations (325) which is thought to influence meiotic recombination frequencies, causes some alterations in euchromatic MR in crosses involving the OKI strain. The other mutation, 318, affects neither meiotic nor dysgenic recombination. Su(var) 325 is the first known "factor" to influence meiotic and dysgenic recombination similarly. / Science, Faculty of / Zoology, Department of / Graduate

Les crossing-overs (CO) sont issus d’échange réciproque de matériel génétique entre les chromosomes homologues. Les COs produisent de la diversité génétique et sont essentiels chez la plupart des eucaryotes, pour la distribution équilibrée des chromosomes lors de la méiose. Malgré leur importance, et un large excès de précurseurs moléculaires, le nombre de CO est très limité dans la grande majorité des espèces (Typiquement 1 à 4 par paire de chromosomes). Cela suggère que les COs sont étroitement régulés, mais les mécanismes sous-jacents sont mal connus. Pour identifier les gènes qui limitent la formation des CO, l’équipe a mené un crible génétique chez Arabidopsis thaliana. Ces travaux ont mené à l’identification de plusieurs facteurs anti-CO, définissant trois voies : (i) L’hélicase FANCM et ses co-facteurs ; (ii) L’AAA-ATPase FIDGETIN-LIKE-1 (FIGL1) ; (iii) Le complexe RECQ4-Topoisomerase 3α-RMI1.Le premier objectif de ma thèse est d’explorer les relations entre ces trois voies en s’attachant aux questions suivantes ; (1) Jusqu’où peut-on augmenter la recombinaison en combinant les mutations dans FANCM, FIGL1 et RECQ4 ? Nous avons montré que la plus forte augmentation de recombinaison était obtenue dans recq4 figl1, atteignant 7,5 fois la fréquence du sauvage en moyenne sur le génome. (2) Quel est la distribution de ces extra-COs ? L’augmentation de recombinaison n’est pas homogène le long du génome : Les fréquences de CO augmente fortement des centromères vers les télomères, avec les plus hautes fréquences observées dans les régions distales. (3) La modification des fréquences de recombinaison est-elle identique lors des méioses mâles et femelle ? Chez le sauvage, la fréquence de recombinaison est plus élevée lors de la méiose mâle que femelle. Au contraire, la recombinaison femelle devient plus élevée que la recombinaison mâle chez les mutants recq4 et recq4 figl1. Ceci suggère que des contraintes qui s’appliquent sur la formation des CO lors de la méiose femelle sont relâchées chez ces mutants. En poursuivant le crible génétique, un nouveau mutant hyper-recombinant a été identifié. Le second objectif de ma thèse fut d’identifier et de caractériser fonctionnellement le gène correspondant. Une cartographie génétique et des études d’interactions protéine-protéine, ont mené à l’identification d’un facteur qui interagit directement avec FIGL1 et semble former un complexe conservé depuis les plantes jusqu’au mammifères. Nous avons baptisé cette protéine FLIP (Fidgetin-like-1 interacting protein). Les fréquences de recombinaisons sont augmentées dans flip-1, confirmant que FLIP1 limite la formation des COs. Des études d’épistasie ont montré que FLIP et FIGL1 agissent dans la même voie. De plus les protéines FIGL1/FLIP d’Arabidopsis ou humaine, interagissent avec RAD51 et DMC1, les deux protéines qui catalyse une étape clef de la recombinaison, l’invasion d’un ADN homologue. Finalement, dans flip comme dans figl1, la dynamique de DMC1 est modifiée. Nous proposons donc un modèle dans lequel le complexe FLIP-FIGL1 régule négativement l’activité de RAD51/DMC1 pour limiter la formation des COs. L’étude du complexe conservé FLIP-FIGL1 a mis en évidence un nouveau mode de régulation de la recombinaison, qui agit vraisemblablement à l’étape clé de l’échange de brin homologue. De plus, l’augmentation des CO sans précédent obtenues chez recq4 figl1 peut être d’un grand intérêt pour l’amélioration des plantes en permettant de diversité de nouvelles combinaisons alléliques. / Meiotic crossovers (CO) are formed by reciprocal exchange of genetic material between the homologous chromosomes. CO generate genetic diversity and are essential for the proper segregation of chromosomes during meiosis in most eukaryotes. Despite their significance and a large excess of CO precursors, CO number is very low in vast majority of species (typically one to three per chromosome pair). This indicates that COs are tightly regulated but the underlying mechanisms of this limit remain elusive. In order to identify genes that limit COs, a genetic screen was performed in Arabidopsis thaliana. This led to the identification and characterization of several anti-CO factors belonging to three different pathways: (i) The FANCM helicase and its cofactors (ii) The AAA-ATPase FIDGETIN-LIKE-1 (FIGL1) (iii) The RECQ4 -Topoisomerase 3α-RMI1 complex. The first objective was to understand the functional relationship between these three pathways and to address following questions: (1) how far can we increase recombination when combining mutations in FANCM, FIGL1 and RECQ4? We show that the highest increase in recombination was obtained in figl1 recq4, reaching to 7.5 fold the wild type level, on average genome wide. (2) How is the distribution of recombination events genome wide in mutants? The increased CO frequency in the mutants was not uniform throughout the genome. CO frequency rises from the centromere to telomeres, with distal intervals having highest COs (3) is the recombination frequency increase same in both male and female? In Arabidopsis wild type, male has higher recombination than female meiosis. In contrast, in recq4 and recq4 figl1, female recombination was higher than male. This suggests that certain constraints that apply to CO formation in wild type females are relieved in the mutant. By continuing the same genetic screen, a novel anti-CO mutant was identified. The second objective was to identify and functionally characterize the corresponding gene. Genetic mapping and protein interaction studies led to the identification of a factor that directly interacts with FIGL1 and appears to form a conserved complex both in Arabidopsis and humans. Hence, the factor was named FLIP (Fidgetin-like-1 interacting protein). Recombination frequency is increased in flip, confirming that FLIP limit COs. Epistasis studies showed that FLIP and FIGL1 act in same pathway. Further, FIGL1/FLIP proteins of Arabidopsis and humans directly interact with the recombinases RAD51 and DMC1 which catalyze a crucial step of homologous recombination, the inter homolog strand invasion. In addition flip like figl1 modifies dynamics of DMC1. We thus propose a model wherein the FLIP-FIGL1 complex negatively regulates RAD51/DMC1 to limit CO formation. Studying the conserved FIGL1-FLIP complex led to the identification of a novel mode of regulation of recombination, that likely acts at the key step of homologous strand invasion. Further the unprecedented level of CO increase in recq4figl1 in hybrids could be of great interest for crop improvement, allowing the production of novel allele combinations.

Meiotic recombination is both a fundamental biological process required for proper chromosomal segregation during meiosis and a fundamental genomic parameter that shapes major features of the genomic landscape. While there is strong evidence of fitness costs of low rates of recombination, the possible fitness costs of high rates of recombination are less defined. To determine whether a single lower fitness bound can explain the variation in recombination rate observed in human populations, we simulated the evolution of recombination rate as a quantitative trait using empirically-derived parameters. For our fitness function, we implemented a hyperbolic tangent curve with flexible parameters to capture a wide range of existing hypotheses. We found that both a lower and upper bound are necessary to explain the observed variation in recombination rate, and we describe a parameter space for an upper bound on recombination rate that produces results consistent with empirical observations.

No description available.

In the past few years, in vivo recombination technologies have emerged to improve the efficiency and simplicity of genetic engineering in Escherichia coli, Salmonella enterica serovar, and other gram-negative bacteria. Phage λ Red homologous recombination system is used to mediate the accurate replacement of target DNA with PCR-generated ?targeting cassettes? that contain flanking regions of shared homologous DNA sequence. However, the efficiency of λ Red-mediated recombineering in Salmonella is far lower than that in Escherichia coli. In this study, I firstly improved the recombineering-based strategy by using linear DNA targeting cassettes that contain long flanking ?arms? of sequence (ca. 1,000 base pairs) homologous to the chromosomal target. This reliable and efficient method enables multiple gene targeting procedures to be performed on a single Salmonella enterica serovar typhi Ty21a (Ty21a) chromosome in a straightforward, sequential manner with high efficiency. Secondly, I applied this improved strategy in construction of Salmonella to be live attenuated oral vaccine and tumor targeting vector. In the first part of this thesis, I describe an improved method in Ty21a. Using this strategy, I inserted three different influenza antigen expression cassettes as well as a green fluorescent protein reporter gene into four different loci on the Ty21a chromosome with high efficiency and accuracy. Fluorescent microscopy and Western blotting analysis confirmed that strong inducible expression of all four heterologous genes could be achieved. The immune response of this vaccine was also evaluated by ELISA and ELIspots. In the second part of this thesisi, I use this improved recombineering strategy to engineer bacteria of Salmonella typhimurium (S. typhimurium) as therapeutic agents against solid tumor. In current study, a major challenge for bacterial therapy of cancer is avoiding damage to normal tissues. Consequently the virulence of bacteria must be adequately attenuated for therapeutic use. An alternative approach was developed here. By placing an essential gene under a hypoxia conditioned promoter, S. typhimurium strain SL7207 was engineered to generate strain YB1 that survives only in anaerobic conditions without otherwise affecting its functions. In breast and liver tumor bearing mice models, YB1 grew within tumor, retarding its growth, while being rapidly eliminated from normal tissues. Mice treated with SL7207 were killed by infection within short period time. Inhibition of tumor growth by YB1 was significant and was enhanced by the addition of 5-FU in breast cancer model. The development of an “obligate” anaerobic Salmonella provides a much safer bacterial vector for further development of anti-tumor therapies without compromising the other functions or tumor fitness of the bacterium as attenuation methods normally do. In summary, I have developed an efficient, robust and versatile method in genome-wide Salmonella genetic manipulation. Furthermore, I used this method to construct a recombinant Ty21a antigen-expressing vaccine strain and a tumor targeting YB1 strain. / published_or_final_version / Biochemistry / Doctoral / Doctor of Philosophy

No description available.

No description available.