Using the citidine deaminase AID as a tool we performed a screening of viable yeast strains deleted in genes with nuclear functions for AID-dependent hyper-recombination. We identified a component of the nuclear pore basket, as a gene involved in preventing R loop accumulation and a DNA helicase, whose role in R loop formation and replication stress was then characterized in depth.
In addition, we performed the following siRNA screenings to search human genes with a role in R loop metabolism: i) the human Druggable Genome siRNA library (4795 siRNAs) by H2AX foci Immunofluorescence (IF) in U2OS stable cell lines for the inducible expression of AID constructed for this study; ii) specific siRNA libraries against the DNA damage response (DDR) genes (240 siRNAs), and nuclear DNA/RNA metabolism related genes (77 siRNAs) in HeLa cells via IF with an anti-DNA-RNA antibody. We validated the positive candidates by DNA-RNA immunoprecipitation (DRIP) assays, selecting 15, including the mitochondrial degradosome subunit or the endonuclease Dicer. This study has led to a new role of the mitochondrial degradosome in preventing harmful R loop formation at mtDNA plus a new role of DDR functions involved in replication stress, including ATR, the 9-1-1 complex and postreplicative repair factors. Other factors with a putative role in R loop processing are being explored by proteomic approaches by expressing a GFP protein fused to RNaseH hybrid-binding domain (HB-GFP) in HEK293 cells.
With respect to functions specifically related to chromatin, we completed the screening and analysis of the Non-Essential Histone H3 & H4 mutant collection of S. cerevisiae for mutants that increased R loops by ectopically expressing the human AID. We identified specific histone mutants that facilitate R loops without causing genomic instability, proving for the first time that R loops are not deleterious per se. The different behavior does not depend on the R loop size but on chromatin modifications such as H3S10-P.
In human cells, we provided evidence that R loop protection by RNA binding proteins may also be promoted by chromatin remodeling. We found that human THO interacts with the Sin3A histone deacetylase complex to suppress R loops, DNA damage, and replication impairment.
Our work supports a role of the SWI/SNF chromatin remodeler complex in helping resolve R-loop-mediated T-R conflicts
We found some RNA helicases as protectors of R loop-mediated instability: UAP56/DDX39B that localizes to active chromatin and prevents the accumulation of RNA–DNA hybrids over the entire genome; and DDX5 that interacts with BRAC2 and promotes resolution of R-loops at DNA breaks facilitating DNA repair.
We analyzed replication fork block and DSB hotspots in yeast R loop-accumulating strains, using hpr1- and sen1- degron strains to induce degradation of Hpr1 and Sen1. Depletion of either protein causes genome-wide analysis of R loops, H3S10P and H2AP distribution that are different, defining DSB hotspots coincident with R loop sites at different cell cycle stages.
Dissemination
Results have been published in top journals (Mol. Cell 2017, PNAS 2017, EMBO J. 2017, Nat Commun. 2018, Genes Dev 2018, EMBO Rep 2019, Cell Rep 2019, Genes Dev 2019, Nature Comm. 2019, Genes Dev 2020, PLoS Genet 2020; EMBO J 2021; Nature Genetics 2021; Nat Commun. 2021 (3 papers); eLife 2021; Nat Commun. 2021) and presented in conferences at top international meetings (Gordon Res. Conf., FASEB, EMBO, NAITO and others) and Universities (Beijing, Busan, Madrid, Viena, Paris, etc).
