DAMAGE-INDUCED LOCALIZED HYPERMUTABILITY Because they use the intact strand as a template after damage removal, the major DNA repair pathways operate on damage in double-strand DNA. Therefore, lesions in transient single-strand stretches of chromosomal DNA are expected to be especially threatening to genome stability. To test this hypothesis, we designed systems in budding yeast that could generate many kb of persistent ssDNA next to double-strand breaks (DSBs) or uncapped telomeres. The systems allowed controlled restoration to the double-strand state after applying DNA damage. We found that lesions induced by UV-light and methyl methanesulfonate can be tolerated in long single-strand regions and are hypermutagenic. The hypermutability required PCNA monoubiquitination and was largely attributable to translesion synthesis by an error-prone DNA polymerase. In support of the idea that multiple lesions in ss DNA can be a source of hypermutability, our analysis of the UV-induced mutants revealed strong strand-specific bias. Importantly, there was an unexpectedly high frequency of alleles with widely-separated multiple mutations scattered over several kb. Since hypermutability and multiple mutations associated with lesions in transient stretches of long ssDNA may be a source of carcinogenesis and provide selective advantage in adaptive evolution, we explored the phenomenon of damage-induced hypermutability further. We addressed the size and continuity of UV-induced hypermutable regions via long-range sequencing, up to 30 kb adjacent to the mutation reporter at an uncapped telomere. We found that up to 15 kb of damaged ssDNA can be efficiently restored to the ds-state even though there are sufficient lesions to cause more than 10 widely-spaced mutations. Furthermore, the entire genomes of several individual colonies arising from UV-irradiated yeast cells were re-sequenced using a Solexa/Illumina platform (collaboration with Dr. Piotr Miezckowski at U. North Carolina, Chapel Hill). We established that at the current level of technology as few as 10 new mutations can be detected in a re-sequenced yeast genome. This puts us in a position to address efficiently genome-wide mutagenesis caused by exogenous and endogenous factors. We found genome-wide subtelomeric clusters of mutations, an increase in overall frequency of subtelomeric mutations and we identified strand bias of mutations in subtelomeric regions. These findings confirm our hypothesis that several long stretches of damaged ssDNA can be efficiently restored to ds-state within the same cell, thereby generating regions of damage-induced hypermutability in multiple genomic locations. In future studies we will utilize the potential of our recently developed novel mutagenesis reporter systems and concentrate on agents of potential significant environmental relevance, especially those where genotoxic effects on dsDNA were found weak, but carcinogenic effects are significant. Of particular interest are agents that may have a synergistic effect at single strand regions. We are addressing if damage-induced localized hypermutability can occur in wild type cells without artificially generated long ssDNA. Using a specially designed screen we found that in the chromosomes of wild type yeast grown on small concentration of alkylation agent (MMS) there were regions of transient hypermutability where up to 30 mutations simultaneously occurred in a 10-100 kb area, while the rest of the genome carried only 10-20 total induced mutations. We plan to continue the study of this new kind of environmental mutagenesis to understand its molecular mechanisms. We will identify genetic and environmental factors (genes, chromosomal location, various kinds of endogenous and exogenous DNA damage, etc.) that keep the frequency of widely-spread multiple mutations at a low level. CHROMOSOMAL DSBS CAN RESULT FROM DNA BASE DAMAGE. We extended our research on base-damage generated DSBs to the finding that accumulation of chromosomal DSB can be caused by mis-routed base excision repair (BER). Since DSBs could be generated during repair of clustered damage, the repair of closely-opposed lesions must be well-coordinated. Using single and multiple mutants of budding yeast that impede interaction of DNA polymerase delta and 5'-flap endonuclease Rad27/Fen1 with the PCNA sliding clamp, we show that the lack of coordination between these components during long-patch base excision repair of alkylation damage can result in many DSBs within the chromosomes of non-dividing haploid cells. This contrasts with the efficient repair of non-clustered MMS-induced lesions, as measured by quantitative PCR and S1 nuclease cleavage of single-strand break sites. We conclude that closely-opposed single-strand lesions are a unique threat to the genome and that repair of closely-opposed strand damage requires greater spatial and temporal coordination between the participating proteins than widely-spaced damage in order to prevent the development of DSBs. Recently we addressed the question if DSBs that we detect in pulsed-field gel electrophoresis analysis are actually formed within cells, rather than chromosome breakage at positions of closely opposed strand lesions during electrophoresis. We hypothesized that if the DSBs are formed in cells, they would serve a substrate for end-resection followed by DSB-repair. To answer this question we took advantage of our recent finding that large chromosome-size DNA molecules with single-strand tails generated by end-resection move slower in PFGE as compared to unresected molecules. We found that in G2-arrested apn1 apn2 (deficient in base-excision repair) yeast cells, MMS-induced DSBs in circular chromosome III are indeed subject to end-resection and repair. We shall extend this approach to G1 cells, in which DSB repair is blocked via a Cdk1/Cdc28 mechanism. We shall explore genetic controls that are known to affect the rate and/or cell cycle control of DSB repair such as RAD55/57 (accessory RAD51-homologues), DOT1 (H3-K79 histone methyltransferase), RAD9 (DSB-sensor checkpoint protein), Ku70/Ku80 (DSB-binding NHEJ proteins). GENE DOSAGE OF GENOME STABILITY GENES. We established a system to measure the effect of reduced expression of genome stability genes on chromosome stability in tetraploid yeast, where gene copy number can be varied from 1 copy (simplex) to 4 (wild type). Surprisingly, a strain with only 1 copy of the cohesin gene MCD1 was much more sensistive to ionizing radiation (IR) and methylmethanesulfonate than strains simplex for the RAD50 and RAD51 genes which are central to much of DSB repair. The MCD1 simplex cells, when arrested at G2/M, are 10 times more sensitive to the same IR dose than WT cells. Moreover, total DSB repair was reduced 4 fold in the MCD1 simplex. In contrast, recombination between homologous chromosomes is enhanced about 10-fold for simplex vs 4 copies. Similar results were found for UV in terms of increased recombination in G2/M cells when the MCD1 copy number was reduced. Also, no significant differences were shown between MCD1 simplex and 4 copies cells when G1 stationary cells were irradiated with IR or UV. Based on these findings, DSB repair processes that are determined by recombinational repair mechanisms are dramatically influenced by the dosage of proteins that are identified with maintaining cohesion between sister chromatids. We also establish that cohesin restricts damage induced recombination to sister chomatids in G2 cells preventing preventing recombination between homologous chromosomes. This cohesin gateway may prevent loss of heterozygosity, which is often associated with cancers in higher organisms.
Showing the most recent 10 out of 20 publications
