Multi-pathway DNA-repair reporters reveal competition between end-joining, single-strand annealing and homologous recombination at Cas9-induced DNA double-strand breaks

Abstract

DNA double-strand breaks (DSB) are repaired by multiple distinct pathways, with outcomes ranging from error-free repair to mutagenesis and genomic loss. DSB-repair pathway cross-talk and compensation is incompletely understood, despite its importance for genomic stability, oncogenesis, and genome editing using CRISPR/Cas9. To address this, we constructed and validated three fluorescent Cas9-based reporters, named DSB-Spectrum, that simultaneously quantify the contribution of multiple DNA repair pathways at a DSB. DSB-Spectrum reporters distinguish between DSB-repair by error-free canonical non-homologous end-joining (c-NHEJ) versus homologous recombination (HR; reporter 1), mutagenic repair versus HR (reporter 2), and mutagenic end-joining versus single strand annealing (SSA) versus HR (reporter 3). Using these reporters, we show that inhibiting the c-NHEJ factor DNA-PKcs increases repair by HR, but also substantially increases mutagenic SSA. Our data indicate that SSA-mediated DSB-repair also occurs at endogenous genomic loci, driven by Alu elements or homologous gene regions. Finally, we demonstrate that long-range end-resection factors DNA2 and Exo1 promote SSA and reduce HR, when both pathways compete for the same substrate. These new Cas9-based DSB-Spectrum reporters facilitate the comprehensive analysis of repair pathway crosstalk and DSB-repair outcome.

EXO1-mediated DNA repair by single-strand annealing is essential for BRCA1-deficient cells

“…In contrast, depletion of these proteins had differential effects on SSA activity (Figure 6D right panel, Supplemental figure 3E). BRCA1 depletion resulted in a decrease in SSA, whereas BRCA2 depletion resulted in an increase in SSA usage, consistent with previous publications (Anantha et al, 2017; Stark et al, 2004; van de Kooij et al, 2022). Importantly, whereas cells lacking both EXO1 and BRCA1 had substantially reduced levels of SSA, concomitant loss of EXO1 did not affect the level of SSA in BRCA2-depleted cells (Figure 6D).…”

“…We thereafter considered that BRCA1-deficient cells depend on the function of EXO1 in long-range end-resection during DSB repair. Using our newly developed multi-pathway reporter for DSB repair, we recently found that long-range end-resection mediated by EXO1 is not required for HR but promotes SSA (van de Kooij et al, 2022). To substantiate these findings in a BRCA1-deficient setting, we examined ionising irradiation-induced foci (IRIF) of RAD51 and RAD52 in our BRCA1-deficient cell lines as a read-out of HR and SSA, respectively.…”

“…We hypothesized that a reliance on EXO1-mediated SSA could explain the specific lethality of EXO1 loss in BRCA1-versus BRCA2-deficient cells. To test this, we depleted BRCA1 or BRCA2 by RNAi in cells containing our multi-pathway reporter for DSB repair that can quantify repair of a Cas9-induced DSB by SSA and HR simultaneously (van de Kooij et al, 2022). The experiments were performed in EXO1-proficient and -deficient reporter cell lines (Supplemental Figure 3E).…”

“…HEK 293T cells were obtained from ATCC (Manassas, VA, USA) and HEK 293T DSB-Spectrum V3 cells were previously described (van de Kooij et al, 2022). H1299 pLKO TetOn sh BRCA2 cells (gift from Dr. Madalena Tarsounas) were previously described (Tacconi et al, 2017).…”

“…In addition to HR, two other DSB repair pathways are also dependent on end-resection: alternative end-joining (alt-EJ) and single-strand annealing (SSA). During alt-EJ, limited end-resection exposes regions of microhomology flanking the DSB-ends which anneal during repair, generally resulting in deletion of the intermittent sequence (Kelso et al, 2019;Mateos-Gomez et al, 2015;van Schendel et al, 2021;Yousefzadeh et al, 2014). SSA requires more extensive long-range end-resection to expose larger stretches of (imperfect) homology which can be located at distal regions up-and downstream of the DSB.…”