Browsing by Subject "Mitotic recombination"

Double-stranded DNA breaks are potentially lethal lesions that can be repaired in mitotic cells by either homologous recombination (HR) or non-homologous end- joining (NHEJ) pathways. In the HR pathway, the broken DNA molecule is repaired using either the sister chromatid or the homolog as a template. Mitotic recombination events involving the homolog often result in loss of heterozygosity (LOH) of markers located distal to the crossover. In humans that are heterozygous for a mutation in a tumor suppressor gene, mitotic recombination leading to LOH can be an early step in cancer development.

In my thesis research, I analyzed mitotic recombination in the yeast Saccharomyces cerevisiae using oligonucleotide-containing microarrays to detect LOH of single-nucleotide polymorphisms (SNPs). In analyzing cells treated with ionizing radiation, I performed the first whole-genome analysis of LOH events done in any organism (Chapter 2). I showed that irradiated cells had between two and three unselected LOH events. I also showed that crossovers were often associated with non- reciprocal exchanges of genetic information (gene conversion events) and that these conversion events were more complex than predicted by standard models of homologous recombination.

In Chapter 3, I describe my mapping of spontaneous crossovers in a 1.1 Mb region of yeast chromosome IV. This analysis is the first high-resolution mitotic recombination map of a substantial fraction (about 10%) of a eukaryotic genome. I demonstrated the existence of recombination "hotspots" and showed that some of these hotspots were homolog-specific. Two of the strongest hotspots were formed by closely- spaced inverted repeats of retrotransposons. I demonstrated that the hotspot activity was a consequence of a secondary DNA structure formed by these repeats. Additionally, the majority of spontaneous LOH events reflect DNA lesions induced in unreplicated chromosomes during G1 of the cell cycle, indicating that G1-initiated lesions threaten genome stability more than G2-initiated lesions.

In Chapter 4, I describe mitotic crossovers associated with DNA replication stress induced by hydroxyurea (HU) treatment. Surprisingly, most HU-induced crossovers had conversion tracts indicative of DNA lesions initiated in G1. Additionally, HU- induced recombination events were very significantly associated with solo delta elements, a 330 bp sequence that is repeated several hundred times in the yeast genome.

In diploid yeast cells, mitotic recombination is very important for repairing double-strand breaks (DSB). When repair of a DSB results in crossovers, it may cause loss of heterozygosity (LOH) of markers centromere-distal to the DSB in both daughter cells. Gene conversion events unassociated with crossovers cause LOH for an interstitial section of a chromosome. Alternatively, DSBs can initiate break-induced replication (BIR), causing LOH in only one of the daughter cells. Mapping mitotic LOH contributes to understanding of mechanisms for repairing DSBs and distribution of these recombinogenic lesions. Methods for selecting mitotic crossovers and mapping the positions of crossovers have recently been developed in our lab. Our current approach uses a diploid yeast strain that is heterozygous for about 55,000 SNPs, and employs SNP-Microarrays to map LOH events throughout the genome. These methods allow us to examine selected crossovers on chromosome V and unselected mitotic recombination events (crossovers, gene conversion events unassociated with crossovers, and BIR events) at about 1 kb resolution across the genome.

Mitotic recombination can be greatly induced by UV radiation. However, prior to my research, the nature of the recombinogenic lesions and the distribution of UV-induced recombination events were relatively uncharacterized. Using SNP microarrays, we constructed maps of UV-induced LOH events in G1-synchronized cells. Mitotic crossovers were stimulated 1500-fold and 8500-fold by UV doses of 1 J/m2 and 15 J/m2, respectively, compared to spontaneous events. Additionally, cells treated with 15 J/m2 have about eight unselected LOH events per pair of sectors, including gene conversions associated and unassociated with crossovers as well as BIR events. These unselected LOH events are distributed randomly throughout the genome with no particular hotspots; however, the rDNA cluster was under-represented for the initiation of crossover and BIR events. Interestingly, we found that a high fraction of recombination events in cells treated with 15 J/m2 reflected repair of two sister chromatids broken at roughly the same position. In cells treated with 1 J/m2, most events reflect repair of a single broken sister chromatid (Chapter 2).

The primary pathway to remove pyrimidine dimers introduced by UV is the nucleotide excision repair (NER) pathway. In NER, the dimer is excised to generate a 30-nucleotide gap that can be replicated to form DSBs if not filled in before DNA replication. The NER gap can also be expanded by Exo1p to form single stranded gaps greater than one kilobase. Alternatively, in the absence of NER, unexcised dimers could result in blocks of DNA replication forks. Resolving the stalled replication fork could lead to recombinogenic breaks. In Chapter 3 and Chapter 4, we analyzed recombination events in strains defective in various steps of processing of UV-induced DNA damage, including exo1 and rad14 mutants.

In Chapter 3, I show that Exo1p-expanded NER gaps contribute to UV-induced recombination events. Interestingly, I also found that Exo1p is also required for the hotspot activity of a spontaneous crossover hotspot involving a pair of inverted Ty repeats. In addition to its role of expanding a nick to a long single-stranded gap, Exo1p is also a major player in DSB end resection. Therefore, I examined the gene conversion tract lengths in strains deleted for EXO1. I found that, although crossover-associated gene conversion tracts become shorter in the exo1 mutant as expected, noncrossover tract lengths remained unaffected. As a result, noncrossover tracts are longer than crossover tracts in the exo1 mutant while the opposite result was observed in the wild-type strains. I proposed models to rationalize this observation.

In Chapter 4, to investigate whether the substantial recombinogenic effect in UV in G1-synchronized cells requires NER, we mapped UV-induced LOH events in NER-deficient rad14 diploids treated with 1 J/m2. Mitotic recombination between homologs was greatly stimulated, which suggests that dimers themselves can also cause recombination without processing by NER. We further show that UV-induced inter-homolog recombination events (noncrossover, crossover and BIR) depend on the resolvase Mus81p, and are suppressed by Mms2p-mediated error-free post-replication repair pathway.

The research described in Chapters, 2, 3, and 4 are in the publications Yin and Petes (2013), Yin and Petes (2014), and Yin and Petes (2015), respectively.

Every living cell faces a multitude of DNA threats in its lifetime because damage to DNA is intrinsic to life itself. A double-strand break (DSB) is the most cytotoxic type of DNA damage and is a potent inducer of chromosomal aberrations. Defects in DSB repair are a major driver of tumorigenesis and are associated with numerous developmental, neurological and immunological disorders. To counteract the deleterious effects of DSBs, organisms have evolved a homologous repair (HR) mechanism that is highly precise. The key to its error-free nature lies in its use of a homologous template in restoring the DSB and its preferential occurrence during late S and G2 phase of the cell cycle when identical sister chromatids are available as templates for repair. However, HR can also engage homologous chromosomes and ectopic substrates that share homology, resulting in mitotic loss-of-heterozygosity (LOH) and unwanted chromosomal aberrations. In this case, understanding of the underlying mechanisms and molecular factors that influence accurate sequence transfer and exchange between two homologous substrates becomes crucial.

The focus of this dissertation is examination of the genetic factors and molecular processes occurring at early intermediate steps (DNA end resection and DNA synthesis) of mitotic recombination in Saccharomyces cerevisiae. To model DSB repair, we established a unique plasmid-based assay with a small 8-base pair (bp) gap in the middle of an 800-bp plasmid substrate. To delineate the molecular structures of strand exchange intermediates during HR, we used a 2% diverged plasmid substrate relative to a chromosomal repair template to generate mismatch-containing heteroduplex DNA (hetDNA) intermediates. The assay was performed in a mismatch repair (MMR)-defective background allowing hetDNA to persist and to segregate into daughter cells at the next round of replication. Unexpectedly, even when MMR was inactivated, sequence analysis of the recombinants revealed patches of gene conversion and restoration reflecting mismatch correction within hetDNA tracts. We showed that, in this system, MMR and nucleotide excision repair (NER) correct mismatches via two different mechanisms. While mispairing of nucleotides triggers MMR, NER is recruited by the subtle 6-methyladenine mark on the plasmid substrate, leading to coincident correction of mismatches. The methylation marks on the plasmid were acquired from the bacterial host’s native restriction-modification system during plasmid propagation.

Formation of hetDNA occurs when a plasmid substrate engages the chromosomal template for repair, forming a D-loop intermediate. D-loop extension requires DNA synthesis by DNA polymerase/s. Translesion synthesis (TLS) polymerases have been implicated in HR in both chicken DT40 cells and fruit fly, but not in yeast. This class of polymerases is known for its low fidelity due to a lack of exonuclease domain and is commonly used for lesion bypass and in extending ends with mismatches. We reported for the first time a requirement of Polζ-Rev1 and Polη (TLS polymerases in S. cerevisiae) for completing gap repair. Moreover, gap-repair efficiency suggested that these two polymerases function independently. We concluded that TLS polymerases are involved in either extending the invading 3’ end and/or in the gap-filling process that completes recombination.

DNA resection of a DSB serves as a primary step to generate a 3’ single-stranded DNA (ssDNA) for subsequent homologous template invasion, but this process has mostly been studied in the absence of a repair template or when downstream HR steps are disabled. To analyze the individual contributions of identified nucleases to DSB resection in the context of repair, we established a chromosomal assay; the substrate size was increased to 4 kilobases (kb) and 85 SNPs were present at ~50 bp intervals. In this chromosomal assay, resection and DNA synthesis influence the length of hetDNA tracts in the final recombinants, allowing these two steps to be analyzed. We specifically focused on synthesis-dependent strand annealing (SDSA) events, where hetDNA reflects DNA synthesis and extent of resection. Our main conclusions are as follows. DNA end resection on the annealing end of NCO products generated by SDSA is not as extensive as one might expect from resection measured in single-strand annealing (SSA) assays. In addition, although the two long-range resection pathways (Sgs1-Dna2 and Exo1) can support recombination in a redundant manner, hetDNA was significantly reduced upon loss of either. End processing of DSBs is predominantly 5’ to 3’, but we also observed loss of sequences (greater than 8 nt but less than 40 nt) at the 3’ termini. We have tested and ruled out the involvement of Mre11 and Polε proofreading activity. Lastly, Pol32 functions as a subunit of Polδ to promote extensive repair synthesis during SDSA. hetDNA tract lengths were significantly shorter in the absence of the Pol32 subunit of Polδ, providing direct evidence that Polδ extends the invading end during HR. Together, this work advances our understanding of how resection nucleases and DNA polymerase/s function to regulate mitotic recombination outcome and influence the molecular patterns of NCOs.

During mitotic cell cycles, DNA experiences many types of endogenous and exogenous damaging agents that could potentially cause double strand breaks (DSB). In S. cerevisiae, DSBs are primarily repaired by mitotic recombination and as a result, could lead to loss-of-heterozygosity (LOH). Genetic recombination can happen in both meiosis and mitosis. While genome-wide distribution of meiotic recombination events has been intensively studied, mitotic recombination events have not been mapped unbiasedly throughout the genome until recently. Methods for selecting mitotic crossovers and mapping the positions of crossovers have recently been developed in our lab. Our current approach uses a diploid yeast strain that is heterozygous for about 55,000 SNPs, and employs SNP-Microarrays to map LOH events throughout the genome. These methods allow us to examine selected crossovers and unselected mitotic recombination events (crossover, noncrossover and BIR) at about 1 kb resolution across the genome. Using this method, we generated maps of spontaneous and UV-induced LOH events. In this study, we explore machine learning and variable selection techniques to build a predictive model for where the LOH events occur in the genome.

Randomly from the yeast genome, we simulated control tracts resembling the LOH tracts in terms of tract lengths and locations with respect to single-nucleotide-polymorphism positions. We then extracted roughly 1,100 features such as base compositions, histone modifications, presence of tandem repeats etc. and train classifiers to distinguish control tracts and LOH tracts. We found interesting features of good predictive values. We also found that with the current repertoire of features, the prediction is generally better for spontaneous LOH events than UV-induced LOH events.