In Vitro Characterization of the Nucleolytic Processing of R-loops

Refvik, Samantha T.
Journal Title
Journal ISSN
Volume Title

DNA damage is a constant threat to genomic integrity, and failure to efficiently and accurately repair damage has been linked to a number of human diseases, including aging, cancer, and neurodegeneration. Complicated repair pathways have evolved to contend with different types of DNA lesions. DNA double-strand breaks (DSBs) are the most lethal form of damage, and are repaired by two major pathways, non-homologous end joining (NHEJ) and homologous recombination (HR). CtBP-Interacting Protein (CtIP) is an essential component of HR, where it is required to noncatalytically stimulate resection, the first step of this pathway. However, CtIP also possesses endonucleolytic activity on 5’ DNA flaps in vitro, and the role for this activity in cells remains unclear. Recently, it was found that depletion of CtIP results in the accumulation of R-loops in human cells, and expression of nuclease-deficient CtIP cannot rescue this phenotype. R-loops are three-stranded nucleic acid structures that can form during transcription, when a nascent RNA strand reanneals back to the template DNA and displaces the nontemplate strand, and at double-strand break sites by currently unknown mechanisms. Further, in cells treated with camptothecin, an inhibitor of topoisomerase I, loss of CtIP nuclease activity results in fewer DNA single-strand breaks (SSBs). From this, we hypothesized that CtIP was nicking R-loop structures to either prevent their formation or promote their resolution. Another DNA repair endonuclease, Xeroderma Pigmentosum Group G (XPG), had previously been implicated in R-loop processing, and its depletion results in both higher R-loops and fewer SSBs after camptothecin treatment as well. Unexpectedly, depletion of both CtIP and XPG resulted in R-loop levels similar to wild-type cells. To explain this, we hypothesized that perhaps R-loops required some processing by either CtIP or XPG to be efficiently recognized by the detection methods used, and that loss of both of these enzymes was preventing R-loop detection, but not necessarily R-loop accumulation. To pursue this hypothesis and better understand the role of CtIP and XPG in processing R-loops, I focused on developing tools for studying R-loop processing in vitro, where the system can be more precisely manipulated and the detection methods more direct. I optimized conditions for creating R-loops by in vitro transcription and by annealing the individual nucleic acid strands and started characterizing how purified human CtIP and XPG act on these substrates. Preliminary data has suggested no CtIP activity on R-loops in vitro, while XPG seems to perform some processing, as previously reported. XPG processing requires further study to confirm if and where XPG is cleaving the R-loop substrate. The lack of CtIP activity, though, may represent a requirement for additional cellular factors that were not included in the in vitro system, or may indicate that CtIP is acting in a different way to prevent R-loop accumulation. Further complicating the existing hypothesis, the antibody S9.6, one of the primary detection methods used to measure R-loop levels in cells, shows no obvious difference in binding affinity between RNA-DNA hybrids and R-loop substrates. This could indicate nuclease processing of R-loops is not a prerequisite to R-loop detection, which would necessitate another explanation for wild-type R-loop levels in cells depleted of both CtIP and XPG. Additional studies will be required to understand if and how CtIP processes R-loops, what the role of CtIP- and XPG-mediated R-loop processing in cells is, and why co-depletion of CtIP and XPG doesn’t result in the same increase in R-loops as either single depletion. Still, the work here optimized several platforms for beginning to answer these questions free from the confounding variables of unknown players and problematic detection methods inherent in cellular studies.