RAD52 and ERCC6L/PICH have a compensatory relationship for genome stability in mitosis.

The mammalian RAD52 protein is a DNA repair factor that has both strand annealing and recombination mediator activities, yet is dispensable for cell viability. To characterize genetic contexts that reveal dependence on RAD52 to sustain cell viability (i.e., synthetic lethal relationships), we performed genome-wide CRISPR knock-out screens. Subsequent secondary screening found that depletion of ERCC6L in RAD52-deficient cells causes reduced viability and elevated genome instability, measured as accumulation of 53BP1 into nuclear foci. Furthermore, loss of RAD52 causes elevated levels of anaphase ultrafine bridges marked by ERCC6L, and conversely depletion of ERCC6L causes elevated RAD52 foci both in prometaphase and interphase cells. These effects were enhanced with combination treatments using hydroxyurea and the topoisomerase IIα inhibitor ICRF-193, and the timing of these treatments are consistent with defects in addressing such stress in mitosis. Thus, loss of RAD52 appears to cause an increased reliance on ERCC6L in mitosis, and vice versa. Consistent with this notion, combined depletion of ERCC6L and disrupting G2/M progression via CDK1 inhibition causes a marked loss of viability in RAD52-deficient cells. We suggest that RAD52 and ERCC6L play compensatory roles in protecting genome stability in mitosis.

Resolution of sequence divergence for repeat-mediated deletions shows a polarity that is mediated by MLH1.

Repeat-mediated deletions (RMDs) are a type of chromosomal rearrangement between two homologous sequences that causes loss of the sequence between the repeats, along with one of the repeats. Sequence divergence between repeats suppresses RMDs; the mechanisms of such suppression and of resolution of the sequence divergence remains poorly understood. We identified RMD regulators using a set of reporter assays in mouse cells that test two key parameters: repeat sequence divergence and the distances between one repeat and the initiating chromosomal break. We found that the mismatch repair factor MLH1 suppresses RMDs with sequence divergence in the same pathway as MSH2 and MSH6, and which is dependent on residues in MLH1 and its binding partner PMS2 that are important for nuclease activity. Additionally, we found that the resolution of sequence divergence in the RMD product has a specific polarity, where divergent bases that are proximal to the chromosomal break end are preferentially removed. Moreover, we found that the domain of MLH1 that forms part of the MLH1-PMS2 endonuclease is important for polarity of resolution of sequence divergence. We also identified distinctions between MLH1 versus TOP3α in regulation of RMDs. We suggest that MLH1 suppresses RMDs with sequence divergence, while also promoting directional resolution of sequence divergence in the RMD product.

The importance of DNAPKcs for blunt DNA end joining is magnified when XLF is weakened.

Canonical non-homologous end joining (C-NHEJ) factors can assemble into a long-range (LR) complex with DNA ends relatively far apart that contains DNAPKcs, XLF, XRCC4, LIG4, and the KU heterodimer and a short-range (SR) complex lacking DNAPKcs that has the ends positioned for ligation. Since the SR complex can form de novo, the role of the LR complex (i.e., DNAPKcs) for chromosomal EJ is unclear. We have examined EJ of chromosomal blunt DNA double-strand breaks (DSBs), and found that DNAPKcs is significantly less important than XLF for such EJ. However, weakening XLF via disrupting interaction interfaces causes a marked requirement for DNAPKcs, its kinase activity, and its ABCDE-cluster autophosphorylation sites for blunt DSB EJ. In contrast, other aspects of genome maintenance are sensitive to DNAPKcs kinase inhibition in a manner that is not further enhanced by XLF loss (i.e., suppression of homology-directed repair and structural variants, and IR-resistance). We suggest that DNAPKcs is required to position a weakened XLF in an LR complex that can transition into a functional SR complex for blunt DSB EJ, but also has distinct functions for other aspects of genome maintenance.

Polθ reverse transcribes RNA and promotes RNA-templated DNA repair.

Genome-embedded ribonucleotides arrest replicative DNA polymerases (Pols) and cause DNA breaks. Whether mammalian DNA repair Pols efficiently use template ribonucleotides and promote RNA-templated DNA repair synthesis remains unknown. We find that human Polθ reverse transcribes RNA, similar to retroviral reverse transcriptases (RTs). Polθ exhibits a significantly higher velocity and fidelity of deoxyribonucleotide incorporation on RNA versus DNA. The 3.2-Šcrystal structure of Polθ on a DNA/RNA primer-template with bound deoxyribonucleotide reveals that the enzyme undergoes a major structural transformation within the thumb subdomain to accommodate A-form DNA/RNA and forms multiple hydrogen bonds with template ribose 2'-hydroxyl groups like retroviral RTs. Last, we find that Polθ promotes RNA-templated DNA repair in mammalian cells. These findings suggest that Polθ was selected to accommodate template ribonucleotides during DNA repair.

XLF acts as a flexible connector during non-homologous end joining.

Non-homologous end joining (NHEJ) is the predominant pathway that repairs DNA double-strand breaks in vertebrates. During NHEJ DNA ends are held together by a multi-protein synaptic complex until they are ligated. Here, we use Xenopus laevis egg extract to investigate the role of the intrinsically disordered C-terminal tail of the XRCC4-like factor (XLF), a critical factor in end synapsis. We demonstrate that the XLF tail along with the Ku-binding motif (KBM) at the extreme C-terminus are required for end joining. Although the underlying sequence of the tail can be varied, a minimal tail length is required for NHEJ. Single-molecule FRET experiments that observe end synapsis in real-time show that this defect is due to a failure to closely align DNA ends. Our data supports a model in which a single C-terminal tail tethers XLF to Ku, while allowing XLF to form interactions with XRCC4 that enable synaptic complex formation.

RNF8 has both KU-dependent and independent roles in chromosomal break repair.

Chromosomal double strand breaks (DSBs) can initiate several signaling events, such as ubiquitination, however the precise influence of such signaling on DSB repair outcomes remains poorly understood. With an RNA interference screen, we found that the E3 ubiquitin ligase RNF8 suppresses a deletion rearrangement mediated by canonical non-homologous end joining (C-NHEJ). We also found that RNF8 suppresses EJ without insertion/deletion mutations, which is a hallmark of C-NHEJ. Conversely, RNF8 promotes alternative EJ (ALT-EJ) events involving microhomology that is embedded from the edge of the DSB. These ALT-EJ events likely require limited end resection, whereas RNF8 is not required for single-strand annealing repair involving extensive end resection. Thus, RNF8 appears to specifically facilitate repair events requiring limited end resection, which we find is dependent on the DSB end protection factor KU. However, we also find that RNF8 is important for homology-directed repair (HDR) independently of KU, which appears linked to promoting PALB2 function. Finally, the influence of RNF8 on EJ is distinct from 53BP1 and the ALT-EJ factor, POLQ. We suggest that RNF8 mediates both ALT-EJ and HDR, but via distinct mechanisms, since only the former is dependent on KU.

BLM has Contrary Effects on Repeat-Mediated Deletions, based on the Distance of DNA DSBs to a Repeat and Repeat Divergence.

Repeat-mediated deletions (RMDs) often involve repetitive elements (e.g., short interspersed elements) with sequence divergence that is separated by several kilobase pairs (kbps). We have examined RMDs induced by DNA double-strand breaks (DSBs) under varying conditions of repeat sequence divergence (identical versus 1% and 3% divergent) and DSB/repeat distance (16 bp-28.4 kbp). We find that the BLM helicase promotes RMDs with long DSB/repeat distances (e.g., 28.4 kbp), which is consistent with a role in extensive DSB end resection, because the resection nucleases EXO1 and DNA2 affect RMDs similarly to BLM. In contrast, BLM suppresses RMDs with sequence divergence and intermediate (e.g., 3.3 kbp) DSB/repeat distances, which supports a role in heteroduplex rejection. The role of BLM in heteroduplex rejection is not epistatic with MSH2 and is independent of the annealing factor RAD52. Accordingly, the role of BLM on RMDs is substantially affected by DSB/repeat distance and repeat sequence divergence.

The canonical non-homologous end joining factor XLF promotes chromosomal deletion rearrangements in human cells.

Clastogen exposure can result in chromosomal rearrangements, including large deletions and inversions that are associated with cancer development. To examine such rearrangements in human cells, here we developed a reporter assay based on endogenous genes on chromosome 12. Using the RNA-guided nuclease Cas9, we induced two DNA double-strand breaks, one each in the GAPDH and CD4 genes, that caused a deletion rearrangement leading to CD4 expression from the GAPDH promoter. We observed that this GAPDH-CD4 deletion rearrangement activates CD4+ cells that can be readily detected by flow cytometry. Similarly, double-strand breaks in the LPCAT3 and CD4 genes induced an LPCAT3-CD4 inversion rearrangement resulting in CD4 expression. Studying the GAPDH-CD4 deletion rearrangement in multiple cell lines, we found that the canonical non-homologous end joining (C-NHEJ) factor XLF promotes these rearrangements. Junction analysis uncovered that the relative contribution of C-NHEJ appears lower in U2OS than in HEK293 and A549 cells. Furthermore, an ATM kinase inhibitor increased C-NHEJ-mediated rearrangements only in U2OS cells. We also found that an XLF residue that is critical for an interaction with the C-NHEJ factor X-ray repair cross-complementing 4 (XRCC4), and XRCC4 itself are each important for promoting both this deletion rearrangement and end joining without insertion/deletion mutations. In summary, a reporter assay based on endogenous genes on chromosome 12 reveals that XLF-dependent C-NHEJ promotes deletion rearrangements in human cells and that cell type-specific differences in the contribution of C-NHEJ and ATM kinase inhibition influence these rearrangements.

Distinct roles of RAD52 and POLQ in chromosomal break repair and replication stress response.

Disrupting either the DNA annealing factor RAD52 or the A-family DNA polymerase POLQ can cause synthetic lethality with defects in BRCA1 and BRCA2, which are tumor suppressors important for homology-directed repair of DNA double-strand breaks (DSBs), and protection of stalled replication forks. A likely mechanism of this synthetic lethality is that RAD52 and/or POLQ are important for backup pathways for DSB repair and/or replication stress responses. The features of DSB repair events that require RAD52 vs. POLQ, and whether combined disruption of these factors causes distinct effects on genome maintenance, have been unclear. Using human U2OS cells, we generated a cell line with POLQ mutations upstream of the polymerase domain, a RAD52 knockout cell line, and a line with combined disruption of both genes. We also examined RAD52 and POLQ using RNA-interference. We find that combined disruption of RAD52 and POLQ causes at least additive hypersensitivity to cisplatin, and a synthetic reduction in replication fork restart velocity. We also examined the influence of RAD52 and POLQ on several DSB repair events. We find that RAD52 is particularly important for repair using ≥ 50 nt repeat sequences that flank the DSB, and that also involve removal of non-homologous sequences flanking the repeats. In contrast, POLQ is important for repair events using 6 nt (but not ≥ 18 nt) of flanking repeats that are at the edge of the break, as well as oligonucleotide microhomology-templated (i.e., 12-20 nt) repair events requiring nascent DNA synthesis. Finally, these factors show key distinctions with BRCA2, regarding effects on DSB repair events and response to stalled replication forks. These findings indicate that RAD52 and POLQ have distinct roles in genome maintenance, including for specific features of DSB repair events, such that combined disruption of these factors may be effective for genotoxin sensitization and/or synthetic lethal strategies.