Double stranded breaks are the most mutagenic kind of DNA damage, and when repaired by the more accurate homologous recombination HR mechanism require DNA resection. DNA resection results in “tailed templates” with a 3’overhang. Work by our lab, and others, recently discovered RNA-DNA hybrids at these resected sites. But how are these RNA-DNA hybrids initiated?
We show that single-stranded DNA (ssDNA) can initiate transcription of RNA-DNA hybrids in the absence of a promoter sequence. While this was long known to be possible in vitro, our results suggest this frequently occurs in S. pombe cells but often remains undetected due to the action of RNaseH enzymes. This single-strand-dependent initiation of RNA-DNA hybrid transcription seems to occur wherever ssDNA is available. One of the prominent sources of ssDNA in cells is R-loops.
R-loops are three-stranded structures that form when RNA invades the double-stranded DNA and hybridises to the complimentary sequence, thus exposing a piece of ssDNA. When an artificial R-loop is created by irreversibly binding an LNA oligo to genomic DNA in vivo this this displaces the ssDNA piece on the opposite strand We show that the displaced ssDNA is transcribed into an RNA-DNA hybrid which extends beyond the LNA binding site. In effect, a double r-loop is formed, where both strands of the dsDNA are hybridised to complimentary RNA. This leads to the exciting possibility that double R-loops are a source of double-stranded RNA because resolving this structure should be energetically neutral. One of the known functions of dsRNA in S.pombe is the establishment of RNAi-dependent heterochromatin.
We suggest that our hypothesis explains links that have previously been observed between R-loops and heterochromatin, and fundamentally changes the understanding of the role of ssDNA in cells. The origin of many dsRNAs in higher eukaryotes is unknown making this model intriguing prospect.