Conservative repair of a chromosomal double-strand break by single-strand DNA through two steps of annealing

Abstract

The repair of chromosomal double-strand breaks (DSBs) is essential to normal cell growth, and homologous recombination is a universal process for DSB repair. We explored DSB repair mechanisms in the yeast Saccharomyces cerevisiae using single-strand oligonucleotides with homology to both sides of a DSB. Oligonucleotide-directed repair occurred exclusively via Rad52- and Rad59-mediated single-strand annealing (SSA). Even the SSA domain of human Rad52 provided partial complementation for a null rad52 mutation. The repair did not involve Rad51-driven strand invasion, and moreover the suppression of strand invasion increased repair with oligonucleotides. A DSB was shown to activate targeting by oligonucleotides homologous to only one side of the break at large distances (at least 20 kb) from the break in a strand-biased manner, suggesting extensive 5' to 3' resection, followed by the restoration of resected DNA to the double-strand state. We conclude that long resected chromosomal DSB ends are repaired by a single-strand DNA oligonucleotide through two rounds of annealing. The repair by single-strand DNA can be conservative and may allow for accurate restoration of chromosomal DNAs with closely spaced DSBs.

FIG. 1.

Systems for generating an inducible and a spontaneous chromosomal DSB. (A) Diagram of the I-SceI-inducible DSB system, presenting the scheme of the CORE-I-SceI cassette inserted into the middle of the TRP5 gene on chromosome VII. The position of the DSB is shown at the I-SceI site. 95-mers e and/or f with homology upstream and downstream from the CORE-I-SceI and the I-SceI site are used to target and repair the break and restore the TRP5 gene. The recombination product generating CORE loss Trp⁺ cells is shown. (B) Chromosome cleavage by I-SceI endonuclease in wild-type (WT) and various mutant strains as determined by pulsed-field electrophoresis. Cells grown in glucose do not induce I-SceI expression, while the expression of I-SceI in galactose results in a DSB. The following strains were tested for cleavage of chromosome VII: FRO-101, wild type, containing a COLD (truncated) I-SceI site; FRO-1, wild type containing a HOT (intact) I-SceI site; and mutant rad51, rad59, and rad52 strains derived from FRO-1. The DSB within TRP5 generates one band of 450 kb and one of 650 kb. Chromosome VII and fragments were detected by Southern hybridization. Each genotype is represented by two immediately adjacent lanes. Based on quantification by ImageQuant software, the broken chromosome VII represents 29% in the wild type, 32 to 33% in rad51, 26 to 27% in rad59, and 37 to 38% in rad52. A similar frequency of DSB formation was also obtained in the BY4742 background after I-SceI cutting within the TRP5 locus (not shown). (C) Diagram of the spontaneous DSB associated with an Alu-IR that generates hairpin-capped ends (for details, see reference 22). 95-mers w and/or c with homology upstream and downstream from the Alu-IR are used to target and repair the break and restore the LYS2 gene. The recombination product generating Lys⁺ cells without the Alu-IR is shown.

FIG. 2.

Oligonucleotide versus homologous chromosome-mediated DSB repair in diploid cells. (A) The diagram shows the position of the CORE-I-SceI cassette and the DSB in TRP5 in one copy of chromosome VII. TRP5 has been replaced with the LEU2 gene in the second copy of chromosome VII. 95-mers e and/or f with homology upstream and downstream from the CORE-I-SceI and the I-SceI break site are used to target and repair the break and restore the TRP5 gene. (B) CORE marker loss (light gray bars) and Trp⁺ colonies (dark gray bars) per 10⁷ viable cells that arise after DSB induction in the TRP5 gene and transformation with e, f, or both (e+f) oligonucleotides or no oligonucleotides (−) in a wild-type (WT) (FRO-922) and in a rad51 homozygous (FRO-923) diploid. The vertical bars correspond to the average values from three to six determinations, and the range (error bars) identifies the standard deviation. The genotype of strains analyzed is indicated above the bars. In order to better compare the results, actual values are provided beneath the bars. Shown is the mean number of “CORE loss” colonies per 10⁷ viable cells, followed by the mean number of Trp⁺ colonies and finally by the percentage of Trp⁺ colonies among all the CORE loss colonies. The possible consequences of a DSB in terms of phenotype are presented on the right side of the graph.

FIG.3.

DSB-mediated stimulation of oligonucleotide targeting to the side of the break in a strand-dependent manner. (A) The diagram shows one copy of chromosome VII, where TRP5 has been replaced with the LEU2 gene (Fig. 2). The second copy of chromosome VII contains a TRP5 locus that has been inactivated by a 31-bp frameshift insertion and a CORE-I-SceI cassette that can generate a DSB either 10 kb upstream or 10 kb downstream of the TRP5::ins31 site. The oligonucleotides e and f can restore the Trp⁺ phenotype. The intact copy of chromosome VII can provide a template for the repair of the DSB as described for the experiments illustrated in Fig. 2. (B) Number of Trp⁺ transformants per 10⁷ viable cells resulting from targeting 1 nmol of oligonucleotide e or f or both following no DSB induction (light gray bars) or induction of the DSB (7 h in galactose) (dark gray bars) that is either 10 kb upstream or 10 kb downstream from TRP5::ins31. The vertical light and dark gray bars correspond to the average values of three determinations; the ranges (error bars) identify the standard deviations. (C) Scheme to address possible resection and strand bias during targeting of the mutated TRP5 locus that is many kilobases (6, 10, and 20 kb downstream or 10 kb upstream) from an I-SceI-induced DSB in yeast haploid cells. The scheme and oligonucleotides are similar to those in panel A above, with the only difference being that repair oligonucleotides R1 and/or R2 were added to repair the DSB. (D) Number of Trp⁺ transformants per 10⁷ viable cells resulting from targeting 0.33 nmol of oligonucleotide e or f to TRP5::ins31 with DSB induction (7 h in galactose) (dark gray bars) and without DSB induction (light gray bars) at a position 6, 10, or 20 kb upstream or 10 kb downstream from TRP5::ins31. Also included were 0.33 nmol of the repair oligonucleotide R1 and 0.33 nmol of R2 to repair the induced DSB. The frequency of CORE marker loss events due to the addition of R1 and R2 was about 1% of viable cells when a DSB was induced and ∼0.001% of viable cells in the absence of DSB induction (not shown). The vertical light and dark gray bars correspond to the average values from three to six determinations; the ranges (error bars) identify the standard deviations. The striped bars show the number of Trp⁺ clones that were without CORE markers, and the percentages of these clones are shown at the right of each bar. Trp⁺ clones that were without CORE were detected only for cells with a DSB transformed with R1 or R2. (E) Number of Trp⁺ clones per 10⁷ viable cells resulting from transformation with oligonucleotide e or f (0.5 nmol) alone or together with the repairing oligonucleotide R1 or R2 (0.5 nmol) after cells were grown in glucose (no DSB; light gray bars) or in galactose for 7 h (DSB; dark gray bars). The DSB was generated 10 kb upstream from TRP5::ins31 (see scheme in panel C). The frequency of CORE marker loss due to the addition of R1 or R2 was about 0.3% of viable cells in which a DSB was induced and ∼0.0002% of viable cells in the absence of DSB induction (not shown). The vertical bars correspond to the average values from four determinations; the ranges (error bars) identify the standard deviations. The striped bars show the number of Trp⁺ clones that were without CORE markers, and the percentages of these clones are shown at the right of each bar. Trp⁺ clones that were without CORE were detected only for cells with a DSB transformed with R1 or R2.

FIG.3.

DSB-mediated stimulation of oligonucleotide targeting to the side of the break in a strand-dependent manner. (A) The diagram shows one copy of chromosome VII, where TRP5 has been replaced with the LEU2 gene (Fig. 2). The second copy of chromosome VII contains a TRP5 locus that has been inactivated by a 31-bp frameshift insertion and a CORE-I-SceI cassette that can generate a DSB either 10 kb upstream or 10 kb downstream of the TRP5::ins31 site. The oligonucleotides e and f can restore the Trp⁺ phenotype. The intact copy of chromosome VII can provide a template for the repair of the DSB as described for the experiments illustrated in Fig. 2. (B) Number of Trp⁺ transformants per 10⁷ viable cells resulting from targeting 1 nmol of oligonucleotide e or f or both following no DSB induction (light gray bars) or induction of the DSB (7 h in galactose) (dark gray bars) that is either 10 kb upstream or 10 kb downstream from TRP5::ins31. The vertical light and dark gray bars correspond to the average values of three determinations; the ranges (error bars) identify the standard deviations. (C) Scheme to address possible resection and strand bias during targeting of the mutated TRP5 locus that is many kilobases (6, 10, and 20 kb downstream or 10 kb upstream) from an I-SceI-induced DSB in yeast haploid cells. The scheme and oligonucleotides are similar to those in panel A above, with the only difference being that repair oligonucleotides R1 and/or R2 were added to repair the DSB. (D) Number of Trp⁺ transformants per 10⁷ viable cells resulting from targeting 0.33 nmol of oligonucleotide e or f to TRP5::ins31 with DSB induction (7 h in galactose) (dark gray bars) and without DSB induction (light gray bars) at a position 6, 10, or 20 kb upstream or 10 kb downstream from TRP5::ins31. Also included were 0.33 nmol of the repair oligonucleotide R1 and 0.33 nmol of R2 to repair the induced DSB. The frequency of CORE marker loss events due to the addition of R1 and R2 was about 1% of viable cells when a DSB was induced and ∼0.001% of viable cells in the absence of DSB induction (not shown). The vertical light and dark gray bars correspond to the average values from three to six determinations; the ranges (error bars) identify the standard deviations. The striped bars show the number of Trp⁺ clones that were without CORE markers, and the percentages of these clones are shown at the right of each bar. Trp⁺ clones that were without CORE were detected only for cells with a DSB transformed with R1 or R2. (E) Number of Trp⁺ clones per 10⁷ viable cells resulting from transformation with oligonucleotide e or f (0.5 nmol) alone or together with the repairing oligonucleotide R1 or R2 (0.5 nmol) after cells were grown in glucose (no DSB; light gray bars) or in galactose for 7 h (DSB; dark gray bars). The DSB was generated 10 kb upstream from TRP5::ins31 (see scheme in panel C). The frequency of CORE marker loss due to the addition of R1 or R2 was about 0.3% of viable cells in which a DSB was induced and ∼0.0002% of viable cells in the absence of DSB induction (not shown). The vertical bars correspond to the average values from four determinations; the ranges (error bars) identify the standard deviations. The striped bars show the number of Trp⁺ clones that were without CORE markers, and the percentages of these clones are shown at the right of each bar. Trp⁺ clones that were without CORE were detected only for cells with a DSB transformed with R1 or R2.

FIG. 4.

Models of rejoining of DSB ends by ssDNA via two SSA events. Annealing interactions between cDNA regions are shown as dotted, red parallel lines; actual pairing is identified as red, parallel lines that are vertical; DNA synthesis is shown as a dotted thick black line with an arrow. (A) Template intermediate model. The first annealing interaction between the single-strand repair oligonucleotide (blue arrow) and the DSB ends is presented as dotted, red parallel lines. The repairing oligonucleotide pairs with the homologous region on the 3′ strand after resection of the 5′ strand. The nonhomologous sequence is clipped away, and DNA synthesis can occur to copy the rest of the oligonucleotide. The original oligonucleotide may be removed by a helicase function, and a second annealing interaction occurs with the other 3′ end of the DSB as shown by the thin, dotted red lines. Pairing of the 3′ strands, followed by clipping of nonhomologous tails and subsequent DNA synthesis and ligation, completes the repair. (B) Bridge intermediate model. The repairing oligonucleotide (blue arrow) anneals through two events with both the 3′ and 5′ unwound ends of a DSB. Cleavage of displaced ends, gap filling, and ligation complete the repair event. The oligonucleotide region is either displaced or ligated after the cleavage of nonhomologous tails and DNA synthesis. (C) Multiple chromosomal breaks or closely spaced DSBs in homologous DNA molecules can be repaired via multiple rounds of SSA in a manner similar to that presented in panel A when strand invasion might not be available or as an alternative to strand invasion. Individual small chromosomal fragments after resection provide opportunities for SSA repair. The multiple breaks might result from closely spaced primary events or even at replication forks during replication of DNA with closely opposed single-strand lesions. (D) Repair of closely opposed DSBs in homologous molecules. Following DSB formation, there is degradation of the 5′ strands allowing an annealing interaction of the complementary single strands. DNA synthesis extends the 3′ strands. Branch migration events, due to equilibrium between the extended 3′ ends and the unwound 5′ tails, trigger reannealing of the extended 3′ strands to their initial chromosomal ends. DNA synthesis and ligation complete the repair of the DSB. Both original homologous molecules are repaired without rearrangements.