Yeast Mre11 and Rad1 proteins define a Ku-independent mechanism to repair double-strand breaks lacking overlapping end sequences

Abstract

End joining of double-strand breaks (DSBs) requires Ku proteins and frequently involves base pairing between complementary terminal sequences. To define the role of terminal base pairing in end joining, two oppositely oriented HO endonuclease cleavage sites separated by 2.0 kb were integrated into yeast chromosome III, where constitutive expression of HO endonuclease creates two simultaneous DSBs with no complementary end sequence. Lack of complementary sequence in their 3' single-strand overhangs facilitates efficient repair events distinctly different from when the 3' ends have a 4-bp sequence base paired in various ways to create 2- to 3-bp insertions. Repair of noncomplementary ends results in a set of nonrandom deletions of up to 302 bp, annealed by imperfect microhomology of about 8 to 10 bp at the junctions. This microhomology-mediated end joining (MMEJ) is Ku independent, but strongly dependent on Mre11, Rad50, and Rad1 proteins and partially dependent on Dnl4 protein. The MMEJ also occurs when Rad52 is absent, but the extent of deletions becomes more limited. The increased gamma ray sensitivity of rad1Delta rad52Delta yku70Delta strains compared to rad52Delta yku70Delta strains suggests that MMEJ also contributes to the repair of DSBs induced by ionizing radiation.

FIG. 1.

HO cleavage sites in the MAT locus of chromosome III in SLY18 and SLY19. After HO endonuclease induction, two complementary breaks are generated in SLY18, whereas in SLY19, noncomplementary breaks will be formed. The locations of primers that were used for PCR amplification and sequence analysis of the repair junctions from survivors are indicated by arrows.

FIG. 2.

Epistasis analyses of the effects of RAD1, RAD52, and YKU70 on gamma-ray hypersensitivity. Logarithmic cultures of yeast cells were harvested and resuspended in phosphate-buffered saline, plated onto YEP-dextrose media, and irradiated with the indicated dose of gamma rays. After 3 to 4 days of growth, surviving colonies were counted, and the survival rate was calculated by comparison to the number of colonies from the mock-treated cells. The percentages of survival of RAD⁺ (solid circles), rad1Δ (open diamonds), yku70Δ (solid diamonds), rad52Δ (open squares), rad52Δ yku70Δ (solid triangles), rad1Δ rad52Δ (open circles), rad14Δ rad52Δ yku70Δ (open triangles), and rad1Δ rad52Δ yku70Δ (solid squares) cells are shown. Each experimental point represents the average of three independent experiments.

FIG. 3.

Model for MMEJ. Shown is the joining event that is the most frequent in SLY19. Putative end processing of 4-bp 3′ noncomplementary overhangs created by the cleavage of two HO endonuclease recognition sequences will expose the microhomology that mediates annealing. Subsequent 3′-flap removal that is mostly dependent on Rad1-Rad10 will prime the fill-in synthesis and ligation by Dnl4. Microhomologies are shown in boldface. Unpaired 3′ flaps are shown as small letters. The gene products that may function at each step are listed to the right of the arrows.