Highly transcribed guanine-run containing sequences, in Saccharomyces cerevisiae, become unstable when topoisomerase I (Top1) is disrupted. Topological changes, such as the formation of extended RNA:DNA hybrids or R-loops or non-canonical DNA structures including G-quadruplexes has been proposed as the major underlying cause of the transcription-linked genome instability. Here, we report that R-loop accumulation at a guanine-rich sequence, which is capable of assembling into the four-stranded G4 DNA structure, is dependent on the level and the orientation of transcription. In the absence of Top1 or RNase Hs, R-loops accumulated to substantially higher extent when guanine-runs were located on the non-transcribed strand. This coincides with the orientation where higher genome instability was observed. However, we further report that there are significant differences between the disruption of RNase Hs and Top1 in regards to the orientation-specific elevation in genome instability at the guanine-rich sequence. Additionally, genome instability in Top1-deficient yeasts is not completely suppressed by removal of negative supercoils and further aggravated by expression of mutant Top1. Together, our data provide a strong support for a function of Top1 in suppressing genome instability at the guanine-run containing sequence that goes beyond preventing the transcription-associated RNA:DNA hybrid formation.

Recombination and mutagenesis are elevated by active transcription. The correlation between transcription and genome instability is largely explained by the topological and structural changes in DNA and the associated physical obstacles generated by the transcription machinery. However, such explanation does not directly account for the unique types of mutations originating from the non-canonical residues, uracil or ribonucleotide, which are also elevated at highly transcribed regions. Based on the previous findings that abasic (AP) lesions derived from the uracil residues incorporated into DNA in place of thymine constitute a major component of the transcription-associated mutations in yeast, we formed the hypothesis that DNA synthesis ensuing from the repair of the transcription-induced DNA damage provide the opportunity for uracil-incorporation. In support of this hypothesis, we show here the positive correlation between the level of transcription and the density of uracil residues in the yeast genome indirectly through the mutations generated by the glycosylase that excise undamaged cytosine as well as uracil. The higher uracil-density at actively transcribed regions is confirmed by the long-amplicon PCR analysis. We also show that the uracil-associated mutations at a highly transcribed region are elevated by the induced DNA damage and reduced by the overexpression of a dUTP-catalyzing enzyme Dut1 in G1- or G2-phases of the cell cycle. Overall, our results show that the DNA composition can be modified to include higher uracil-content through the non-replicative, repair-associated DNA synthesis. Uracil in DNA, a major source of spontaneous mutations, can occur through the deamination of cytosine residues or through the direct incorporation of dUTP by DNA polymerases. Recent studies in yeast have shown that the uracil-associated mutations occur more frequently at highly transcribed regions. Because the reduction in dUTP pool decreased these mutations, it was postulated that the extent of uracil-incorporation into DNA is significantly affected by the local transcription activity. We show here that the higher transcription rate does correlate with the higher uracil-density in the yeast genome. We further provide multiple lines of evidence supporting a model of uracil-incorporation into DNA that is dependent on the repair synthesis of transcription-associated DNA damage.

Non-canonical residue in DNA is a major and conserved source of genome instability. The appearance of uracil residues in DNA accompanies a significant mutagenic consequence and is regulated at multiple levels, from the concentration of available dUTP in the nucleotide pool to the excision repair for removal from DNA. Recently, an interesting phenomenon of transcription-associated elevation in uracil-derived mutations was described in Saccharomyces cerevisiae genome. While trying to understand the variability in mutagenesis, we uncovered that the frequency of uracil incorporation into DNA can vary depending on the transcription rate and that the non-replicative, repair-associated DNA synthesis underlies the higher uracil density of the actively transcribed genomic loci. This novel mechanism brings together the chemical vulnerability of DNA under transcription and the uracil-associated mutagenesis, and has the potential to apply to other non-canonical residues of mutagenic importance.

The comet assay is a versatile assay for detecting DNA damage in eukaryotic cells. The assay can measure the levels of various types of damage, including DNA strand breaks, abasic sites and alkali-sensitive sites. Furthermore, the assay can also be modified to include purified DNA glycosylases so that alkylated and oxidized bases can be detected. The CometChip is a higher throughput version of the traditional comet assay and has been used to study cultured cells. Here, we have tested its utility for studies of DNA damage present in vivo. We show that the CometChip is effective in detecting DNA damage in multiple tissues of mice exposed to the direct-acting methylating agent methylmethane sulfonate (MMS) and to the metabolically activated methylating agent N-nitrosodimethylamine (NDMA), which has been found to contaminate food, water, and drugs. Specifically, results from MMS-exposed mice demonstrate that DNA damage can be detected in cells from liver, lung, kidney, pancreas, brain and spleen. Results with NDMA show that DNA damage is detectable in metabolically competent tissues (liver, lung, and kidney), and that DNA repair in vivo can be monitored over time. Additionally, it was found that DNA damage persists for many days after exposure. Furthermore, glycosylases were successfully incorporated into the assay to reveal the presence of damaged bases. Overall, this work demonstrates the efficacy of the in vivo CometChip and reveals new insights into the formation and repair of DNA damage caused by MMS and NDMA.