Rad1, rad10 and rad52 mutations reduce the increase of microhomology length during radiation-induced microhomology-mediated illegitimate recombination in saccharomyces cerevisiae

Abstract

Abstract Illegitimate recombination can repair DNA double-strand breaks in one of two ways, either without sequence homology or by using a few base pairs of homology at the junctions. The second process is known as microhomology-mediated recombination. Previous studies showed that ionizing radiation and restriction enzymes increase the frequency of microhomology-mediated recombination in trans during rejoining of unirradiated plasmids or during integration of plasmids into the genome. Here we show that radiation-induced microhomology-mediated recombination is reduced by deletion of RAD52, RAD1 and RAD10 but is not affected by deletion of RAD51 and RAD2. The rad52 mutant did not change the frequency of radiation-induced microhomology-mediated recombination but rather reduced the length of microhomology required to undergo repair during radiation-induced recombination. The rad1 and rad10 mutants exhibited a smaller increase in the frequency of radiation-induced microhomology-mediated recombination, and the radiation-induced integration junctions from these mutants did not show more than 4 bp of microhomology. These results suggest that Rad52 facilitates annealing of short homologous sequences during integration and that Rad1/Rad10 endonuclease mediates removal of the displaced 3' single-stranded DNA ends after base-pairing of microhomology sequences, when more than 4 bp of microhomology are used. Taken together, these results suggest that radiation-induced microhomology-mediated recombination is under the same genetic control as the single-strand annealing apparatus that requires the RAD52, RAD1 and RAD10 genes.

FIG. 1

The distribution of microhomology use in spontaneous and radiation-induced NHI events in (panel A) wild-type strain RSY12 [data from ref. (20)] and is shown here for comparison. Panel B: rad51 and (panel C) rad52 mutants, in which 10 Gy to the rad52 mutant is an equitoxic dose to 50 Gy in wild-type cells.

FIG. 2

Target sequences of NHI events in rad51 mutant. Panel A: Spontaneous events; panel B: after exposure to radiation. The genomic location, locus and gene of the target sites are shown in the Supplementary Material, Table S1.

FIG. 3

Target sequences of NHI events in rad52 mutant. Panel A: Spontaneous events; panel B: after exposure to 50 Gy of γ radiation; panel C: after exposure to 10 Gy of γ radiation. The genomic location, locus and gene of the target sites are shown in Table S2.

FIG. 4

The distribution of microhomology use in spontaneous and radiation-induced NHI events in (panel A) rad1, (panel B) rad10 and (panel C) rad2 mutants.

FIG. 5

Target sequences of NHI events in rad1 mutant. Panel A: Spontaneous events; panel B: after exposure to γ radiation. The genomic location, locus and gene of the target sites are shown in Table S3.

FIG. 6

Target sequences of NHI events in rad10 mutant. Panel A: Spontaneous events; panel B: after exposure to γ radiation. The genomic location, locus and gene of the target sites are shown in Table S4.

FIG. 7

Target sequences of NHI events in rad2 mutant. Panel A: Spontaneous events; panel B: after exposure to γ radiation. The genomic location, locus and gene of the target sites are shown in Table S5.

FIG. 8

The actual effect of each mutation on MHMR and non-MHMR events after normalization with the relative frequency of NHI. Panel A: The normalized frequency of MHMR events. Panel B: The normalized frequency of non-MHMR events. The relative frequency of NHI resulted from each transformation (the mean shown in Table 1) was multiplied by the fraction of MHMR or non-MHMR events among the nonhomologous integrants isolated from each transformation. The average normalized frequencies of MHMR or non-MHMR were shown in the graph. The normalized frequencies of MHMR or non-MHMR events in different mutants were compared by Student’s t test. *P < 0.05, **P < 0.01, ***P < 0.005.

FIG. 9

Schematic diagram of the mechanism of spontaneous and radiation-induced MHMR. In unirradiated cells, the majority of spontaneous nonhomologous integration events are mediated by illegitimate recombination that does not use microhomology. In irradiated cells, the MHMR pathway is induced to facilitate illegitimate recombination that is probably in trans. Rad52 mediates microhomology search and annealing between short homologous sequences of one end of the integrating DNA and the genomic sequences (shown by base-pairing). The Rad1/Rad10 complex is required for cleavage of the 3′ displaced single strand from the integrating DNA when more than 4 bp of microhomology is involved during integration. The bold arrows indicate the dominant mechanism.