Induction of homologous recombination by site-specific replication stress

Abstract

DNA replication stress is one of the primary causes of genome instability. In response to replication stress, cells can employ replication restart mechanisms that rely on homologous recombination to resume replication fork progression and preserve genome integrity. In this review, we provide an overview of various methods that have been developed to induce site-specific replication fork stalling or collapse in eukaryotic cells. In particular, we highlight recent studies of mechanisms of replication-associated recombination resulting from site-specific protein-DNA barriers and single-strand breaks, and we discuss the contributions of these findings to our understanding of the consequences of these forms of stress on genome stability.

Keywords: Double-strand break (DSB); Homologous recombination; Rad51; Replication restart; Replication stress.

Figure 1.. Model for HR induced by a protein-DNA barrier.

Upon replication fork stalling at a protein-DNA barrier (STOP sign), the replication fork can reverse, providing an end for nuclease mediated 5′-3′ resection. The 3′ end of the reversed fork can then invade the reannealed parental strands, resulting in replication restart. Recombination-dependent synthesis is insensitive to the barrier [26]. Alternatively, nucleases could degrade the nascent lagging strand prior to fork reversal to generate a ssDNA gap that could anneal with the other template strand, displacing the nascent leading strand for Rad51 assembly. At some protein-induced fork barriers, DSBs are detected, suggesting cleavage of the stalled or reversed fork by a structure-selective nuclease to create a broken replication fork. Invasion of the sister chromatid by the broken arm is required to restart replication. The extent of synthesis in the context of the D-loop could be limited by an incoming replication fork or could continue to the telomere via a BIR-like mechanism. Newly synthesized DNA is shown in light blue, while the original template strands at the start of S-phase are shown in dark blue. Recombination-dependent DNA synthesis is indicated by a dotted light blue line.

Figure 2.. Genetic reporters to detect sister-chromatid recombination.

A. Direct repeat ade6 reporter used to detect HR at stalled replication forks in S. pombe [64]. Two configurations of the reporter are shown, one with the fork block upstream (left) and the other with the RTS1 site between the repeats (right). Invasion of the reversed fork into the downstream ade6 allele can generate Ade⁺ recombinants after fork regression, convergence with a centromeric replication fork or by cleavage of the D-loop intermediate. When RTS1 is upstream of the reporter, a template switch is required to generate recombinants [65]. B. The reporter used in mouse cells consists of a full-length copy of GFP that is interrupted by an I-SceI cut site and Ter repeats or an FRT site, and a truncated copy of GFP (tr-GFP) to serve as a donor during repair [82, 85, 113]. gRNAs that target the I-SceI cut-site sequence can be used with Cas9 or nCas9 to generate a DSB or nick, respectively [124, 126]. The GFP repeats are separated by fragments of RFP that yield functional RFP after RNA splicing of the LTGC product. If the fork breaks at the Ter repeats or collapses at a nick, the resulting end can invade Tr-GFP and by short tract synthesis generate a GFP⁺ product (STGC) or GFP⁺ RFP⁺ if synthesis extends from the first to second copy of GFP on the sister chromatid (LTGC). Some LTGC products terminate by NHEJ or MMEJ instead of using the GFP homology (not shown). GFP⁻RFP⁺ products (TDs) are of variable size with microhomologies at the breakpoints.

Figure 3.. Model for replication fork progression/collapse at an SSB.

For a leading strand nick, CMG is most likely lost from the leading strand resulting in a seDSB (not shown). For a lagging strand nick, a seDSB could be generated if the replicative helicase translocates onto the parental dsDNA and is then ubiquitylated for removal from chromatin by p97 [102]. The seDSB is predicted to invade into the intact sister chromatid for replication restart. The D-loop could be resolved by a converging fork generating a single HJ intermediate or synthesis could proceed to the telomere. Alternatively, the seDSB could be converted to deDSB by an incoming fork and be repaired by similar mechanisms to a canonical DSB. A deDSB could also be generated if CMG continues translocation on the leading strand and lagging strand synthesis initiates to the site of the nick. One end of the resulting deDSB is predicted to be blunt and the other to have a ssDNA gap up to the size of an Okazaki fragment [102]. Newly synthesized DNA is shown in light blue, while the original template strands at the start of S-phase are shown in dark blue. Recombination-dependent DNA synthesis is indicated by a dotted light blue line.