Homologous Recombinational Repair of Double-Strand Breaks in Yeast Is Enhanced by MAT Heterozygosity Through yKU-Dependent and -Independent Mechanisms

Corresponding author: Jac A. Nickoloff, Department of Molecular Genetics and Microbiology, University of New Mexico School of Medicine, Albuquerque, NM 87131., jnickoloff{at}salud.unm.edu (E-mail)

Communicating editor: L. S. SYMINGTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

DNA double-strand breaks (DSBs) are repaired by homologous recombination (HR) and nonhomologous end-joining (NHEJ). NHEJ in yeast chromosomes has been observed only when HR is blocked, as in rad52 mutants or in the absence of a homologous repair template. We detected yKu70p-dependent imprecise NHEJ at a frequency of 0.1% in HR-competent Rad⁺ haploid cells. Interestingly, yku70 mutation increased DSB-induced HR between direct repeats by 1.3-fold in a haploid strain and by 1.5-fold in a MAT homozygous (a/a) diploid, but yku70 had no effect on HR in a MAT heterozygous (a/) diploid. yku70 might increase HR because it eliminates the competing precise NHEJ (religation) pathway and/or because yKu70p interferes directly or indirectly with HR. Despite the yku70-dependent increase in a/a cells, HR remained 2-fold lower than in a/ cells. Cell survival was also lower in a/a cells and correlated with the reduction in HR. These results indicate that MAT heterozygosity enhances DSB-induced HR by yKu-dependent and -independent mechanisms, with the latter mechanism promoting cell survival. Surprisingly, yku70 strains survived a DSB slightly better than wild type. We propose that this reflects enhanced HR, not by elimination of precise NHEJ since this pathway produces viable products, but by elimination of yKu-dependent interference of HR.

DNA double-strand breaks (DSBs) can be repaired by homologous recombination (HR) or nonhomologous end-joining (NHEJ). It is thought that HR is the dominant repair mode in the yeast Saccharomyces cerevisiae, while NHEJ plays a larger role in mammalian cells. There are several distinct modes of HR, including conservative processes such as gene conversion and crossing over, and the nonconservative process termed single-strand annealing (SSA) that operates between direct repeats. Gene conversion involves nonreciprocal transfer of continuous blocks of information from a donor to a recipient allele, termed a conversion tract. Conversion tract lengths reflect both heteroduplex DNA (hDNA) formation, resulting from strand invasion and branch migration of Holliday junctions, and mismatch repair of hDNA (PETES et al. 1991 ; NICKOLOFF and HOEKSTRA 1998 ; WENG and NICKOLOFF 1998 ; NICKOLOFF et al. 1999 ). NHEJ involves interactions between regions sharing little or no homology. NHEJ can be nonconservative and mutagenic since ends can be joined imprecisely via annealing between single-stranded ends sharing short (1–5 bp) homologies. DSBs with cohesive ends, such as those generated by endonucleases, can be repaired by conservative, precise NHEJ (religation).

In yeast, most DSB-induced HR requires RAD52 and other members of the RAD52 epistasis group (PAQUES and HABER 1999 ). NHEJ is Rad52p independent, but instead requires yKu70p and yKu80p (which forms the yKu heterodimer) and involves the Rad50p-Mre11p-Xrs2p complex, Lif1p, and ligase IV (CRITCHLOW and JACKSON 1998 ). yKu70p also serves a DNA end protection function since yku70 mutants process DSBs to yield longer 3' single-stranded tails than wild type (LEE et al. 1998 ). Recent studies have shown that NHEJ levels are influenced by mating-type status. Haploid cells, expressing either MATa or MAT, and diploids homozygous at MAT, have levels of NHEJ 10-fold higher than those of cells expressing both a and (e.g., a/ diploids or haploid Sir^- mutants; ASTROM et al. 1999 ; LEE et al. 1999 ). Mating-type heterozygosity enhances DNA repair and HR (FRIIS and ROMAN 1968 ; HEUDE and FABRE 1993 ; FASULLO and DAVE 1994 ; FASULLO et al. 1999 ; LEE et al. 1999 ), but it has not been clear how much of this effect was due to downregulation of the competing NHEJ pathway (yKu dependent) and how much was yKu independent.

Because of the high efficiency of DSB repair by RAD52-dependent HR, prior strategies for detecting NHEJ in yeast chromosomes employed rad52 mutants (KRAMER et al. 1994 ; MOORE and HABER 1996B ) or systems in which a broken molecule had no homologous repair template (SCHIESTL and PETES 1991 ; SCHIESTL et al. 1993 ; MANIVASAKAM and SCHIESTL 1998 ). Although these studies clearly indicate that HR is much more efficient than NHEJ in yeast, they did not provide estimates of the relative rates of DSB repair via HR and NHEJ in strains fully competent to carry out HR (i.e., in Rad⁺ cells suffering a DSB in a duplicated region).

Here we report measures of the relative rates of repair of HO nuclease-induced DSBs by NHEJ and HR in Rad⁺ HR-competent haploid and diploid yeast. We detected imprecise NHEJ in haploid cells at a frequency of 0.1%. HR was increased by yku70 mutation and by MAT heterozygosity. Part of the increase in HR seen with MAT heterozygosity was yKu dependent, but the majority was yKu independent, and the latter correlated with increased cell survival. We made the surprising finding that yku70 mutation slightly increases cell survival following a DSB; this result is discussed in relation to possible mechanisms by which yku70 mutation enhances HR.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plasmid DNA and yeast strains:
Plasmid preparation and manipulation and yeast culture were described previously (SAMBROOK et al. 1989 ; SWEETSER et al. 1994 ; TAGHIAN and NICKOLOFF 1996 ; CHO et al. 1998 ). Strain JW3082, with ura3 direct repeats flanking LEU2 and pUC19, was described previously (CHO et al. 1998 ). The left (donor) ura3 allele was inactivated by a +1 frameshift mutation (X764); the right (recipient) allele was inactivated by an HO site insertion into NcoI (HO432) and contained nine phenotypically silent restriction fragment length polymorphism (RFLP) mutations. JW3082 has a MATa-inc mutation to prevent HO cleavage of MAT and subsequent mating-type interconversion and diploidization (SWEETSER et al. 1994 ). Strain DY3515-13 is a diploid with the same ura3 alleles as JW3082 present in an allelic configuration (NICKOLOFF et al. 1999 ). These recombination substrates are diagrammed in Fig 1. JW3082 and DY3515-13 carry GALHO to allow delivery of DSBs to HO sites when cells are grown in medium with galactose. yku70 mutant strains were constructed by transformation with XmnI-digested plasmid pAF1 (SIEDE et al. 1996 ) (kindly provided by Anna Friedl). This replaces the endogenous YKU70 locus with TRP1-disrupted yku70; mutant status was confirmed by Southern hybridization, growth defects at 37° (SIEDE et al. 1996 ), and reduced efficiencies of transformation with a linearized HIS3/ARS1/CEN4 plasmid (data not shown).

View larger version (17K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Recombination substrates. (Top) ura3 direct repeats separated by pUC19 and LEU2 in JW3082 (CHO et al. 1998 ) and the yku70 derivative GJK3465. The left copy is inactivated by X764 but is otherwise wild type, and the right copy is inactivated by insertion of an HO site at NcoI (HO432) and contains nine silent RFLP markers (shading); the RFLP markers were not scored in the present study. (Bottom) The same ura3 genes present in allelic positions at the normal chromosome V position in DY3515-13 (NICKOLOFF et al. 1999 ) and its a/a and yku70 derivatives. The flanking pUC19 and LEU2 sequences were introduced during construction; the allelic substrates are not flanked by linked repeats.

Diploid strains constructed from MATa-inc and MAT haploids were converted to MATa-inc/MATa-inc by 2-hr expression of GALHO. Cells were then plated for single colonies on YPD; 50% were a-maters (either MATa-inc/MATa-inc or MATa-inc/MATa), and most had no changes in ura3. We confirmed that a-mating strains were MATa-inc/MATa-inc as they did not switch to nonmaters upon induction of GALHO. Genotypes of yeast strains are given in Table 1.

Recombination assays:
DSB-induced and uninduced recombination frequencies were measured using nonselective assays (CHO et al. 1998 ). Briefly, 2-day-old colonies of parent strains were inoculated into 1.5 ml of YPGly medium and incubated for 24 hr. Cultures were divided, and cells were harvested by centrifugation, suspended in 1.5 ml of YPD (uninduced control) or YPGal (with 2% galactose; HO nuclease-induced), grown for 6 hr, and plated on YPD medium. JW3082 recombinants have one of four phenotypes (Fig 2). Ura⁺ Leu⁺ (gene conversion + unequal exchange), Ura⁺ Leu^- (deletion), and Ura^- Leu^- (deletion) products were identified by replica-plating to appropriate media. Ura^- Leu⁺ recombinants and parental cells have the same phenotypes, but these can be distinguished in a replica-plate assay involving reinduction of HO nuclease (WENG et al. 1996 ; CHO et al. 1998 ). In this assay, induction of HO stimulates HR in parental cells since these retain the HO site (producing many Ura⁺ papillae in each colony transferred to uracil omission medium), whereas Ura^- Leu⁺ recombinants, which lack HO432 and are homozygous X764, do not yield Ura⁺ papillae. Among Ura^- Leu⁺ recombinants, HO site loss reflects either long-tract gene conversion, which coconverts X764 (homozygous X764 and homozygous NcoI at position 432), or HO site inactivation by imprecise NHEJ yielding deletions or insertions (heterozygous at both X764 and NcoI). Primers complementary to a sequence downstream of ura3 (5'-TGGAGTTCAATGCGTCCAT-3') and the 3' end of the LEU2 fragment (5'-GGCACCACACAAAAAGTT-3') were used to amplify a 1.3-kbp fragment containing the recipient ura3 allele by PCR. Digestion of PCR products with NcoI identified gene conversions since these convert HO432 to NcoI. NcoI-resistant products were usually imprecise NHEJ products; some retained HO432 and presumably reflect inactivation of HO nuclease or the galactose regulatory system (GALHO^-). Ura^- Leu^- products could arise by HR (crossover, SSA, or unequal sister chromatid exchange) or by NHEJ, and these events were distinguished by Southern hybridization. Junctions formed by NHEJ were identified by DNA sequencing of rescued alleles as described (CHO et al. 1998 ) or by direct sequencing of PCR products.

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Types of DSB repair products for ura3 direct repeats. The parent structure is shown at the top. Three classes of events give rise to four phenotypes, distributed among six main product types. Short- and long-tract gene conversion yields heterozygous and homozygous X764, respectively. Triplications resulting from unequal sister chromatid exchange (Ura+ Leu+) are rare (not shown). All popouts are Leu-. Imprecise NHEJ (Ura- Leu+) may delete some or all of HO432, indicated by HO*; larger deletions from NHEJ (Ura- Leu-) may remove some or all of the right ura3 gene and some or all of LEU2, and may extend further (bottom product).

DY3515-13 recombinants are Ura⁺ or Ura^-, identified using uracil omission media and reinduction assays, respectively. NHEJ products of DY3515-13 were first sought among 130 Ura^- products by using a PCR/NcoI screen as above, except that both copies of ura3 were amplified. An additional 730 Ura^- products were screened by using a pooling approach as follows. Ura^- products were grown to stationary phase in 5 ml of YPD, and 73 pools were made by mixing 0.5-ml aliquots of each of 10 products. PCR was used to amplify both copies of ura3 from genomic DNA isolated from each pool, and PCR products were analyzed by Southern hybridization using a ³²P-labeled probe specific to the wild-type URA3 sequence opposite X764 (5'-TTTTGTTATCGGCTT-3'). This probe hybridizes to X764 heterozygotes (imprecise NHEJ or GALHO^-), but not to X764 homozygotes (gene conversion). A reconstruction experiment indicated that this strategy reliably detects a single heterozygote in a pool with nine X764 homozygotes. All products from each pool that had one or more X764 heterozygotes were retested individually by the PCR/Southern assay to identify X764 heterozygotes. X764^- alleles were rescued as described (NICKOLOFF et al. 1999 ) and sequenced to distinguish imprecise NHEJ and GALHO^- products. Complete product independence was guaranteed for putative NHEJ products since at most one candidate was characterized from each population of parent strains. Statistical analyses were performed with t-tests unless otherwise specified.

Measurement of DSB levels:
DSBs were quantified essentially as described previously (WENG et al. 2000 ). Briefly, HO nuclease was induced for 4 or 6 hr and genomic DNA was prepared. For haploid strains, DSBs were detectable at 4- and 6-hr time points; we present the 4-hr data as this is least likely to be affected by repair. For diploid strains, DSBs were barely detectable at 4 hr, so only the 6-hr data are shown. For haploid strains, HindIII-digested genomic DNA was probed with a ³²P-labeled ura3 fragment consisting of a 0.8-kbp sequence 3' of HO432; this detects a 1.2-kbp donor fragment and a 6-kbp recipient fragment. Upon induction of HO nuclease, the 6-kbp fragment is cleaved into two fragments, but the probe detects only the smaller (0.8-kbp) fragment. DSB levels were calculated as the ratio of the signal from the 0.8-kbp fragment to the sum of the signals from all hybridizing fragments [quantified using a Molecular Dynamics (Sunnyvale, CA) Phosphorimager]. An analogous Southern strategy was used to measure DSB levels in diploid strains.

Cell survival and mating-type switching:
Cell survival was assessed by measuring plating efficiency (PE) following 6 hr galactose induction or 6 hr growth in glucose as a control. PE was calculated as the ratio of YPD colonies to the number of cells plated. Cell numbers were determined using a Coulter Counter, and 350–1600 YPD colonies were scored per determination. Mating-type switching (from MATa-inc/MAT to MATa-inc/MATa-inc or to MATa-inc/MATa) was stimulated by using standard GALHO-induction conditions described above. Cells were plated on YPD after 0, 2, 4, or 6 hr of growth in galactose medium and incubated for 2 days; colonies that had switched mating type were identified as those able to mate with a MAT strain.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Experimental design:
We examined relative rates of DSB repair by HR and NHEJ in Rad⁺ haploid and diploid yeast strains with direct repeat and allelic recombination substrates (Fig 1). Because yKu70p plays a key role in NHEJ, we also examined DSB repair in isogenic yku70 strains. Strain JW3082 (CHO et al. 1998 ) and its yku70 derivative (GJK3465) carry ura3 direct repeats flanking pUC19 and LEU2. One copy of ura3 was inactivated by a +1 frameshift mutation (X764). The second copy was inactivated by an HO site insertion (HO432) and contained nine phenotypically silent RFLP mutations. Diploid strain DY3515-13 (NICKOLOFF et al. 1999 ) and its derivatives have these same ura3 genes at allelic positions. All strains have a copy of GALHO integrated at lys2, providing a galactose-regulated source of HO nuclease to deliver DSBs to HO432. JW3082 has a MATa-inc mutation to prevent HO cleavage of MAT and subsequent mating-type interconversion and diploidization (SWEETSER et al. 1994 ). The MATa-inc mutation is a single-base change that does not affect MAT coding potential; hence, in this report MATa-inc and a are equivalent.

HR in JW3082 can yield products with one of four phenotypes (Fig 2). Most Ura⁺ Leu⁺ products reflect short-tract gene conversion, which conserves the gross structure of the direct repeat; Ura⁺ Leu⁺ products may also result from unequal sister chromatid exchange, yielding three copies of ura3 and two copies of LEU2, but these are rare in JW3082 and related strains (CHO et al. 1998 ; NICKOLOFF et al. 1989 ). Ura⁺ Leu^- and Ura^- Leu^- products ("popouts") reflect loss of pUC19, LEU2, and one copy of ura3 by crossover, SSA, or unequal sister chromatid exchange (RAY et al. 1988 ; NICKOLOFF et al. 1989 ; FISHMAN-LOBELL et al. 1992 ). Most Ura^- Leu⁺ products arise by long-tract gene conversion in which HO432 and X764 coconvert. Precise NHEJ restores the parental structure, but imprecise NHEJ can yield small deletions or insertions that inactivate the HO site (Ura^- Leu⁺) or large deletions (>900 bp) extending from HO432 into the LEU2 coding sequence (Ura^- Leu^-). With the allelic substrates in strain DY3515-13 and its derivatives, Ura⁺ products reflect short-tract gene conversion, and Ura^- products reflect either long-tract conversion extending past X764 or imprecise NHEJ. We showed previously that GALHO induction in JW3082 and DY3515-13 increases HR by >100-fold (CHO et al. 1998 ; NICKOLOFF et al. 1999 ), and similar results were obtained in our study (data not shown). These induction levels ensure that essentially all products analyzed were DSB induced.

DSB repair by imprecise NHEJ in haploid, Rad⁺, HR-competent yeast yields small deletions and insertions and requires YKU70:
Imprecise NHEJ in yeast chromosomal DNA had previously been observed only in strains defective in HR, such as rad52 mutants, or in the absence of a homologous repair template (SCHIESTL and PETES 1991 ; SCHIESTL et al. 1993 ; KRAMER et al. 1994 ; MOORE and HABER 1996B ; MANIVASAKAM and SCHIESTL 1998 ). To detect imprecise NHEJ in haploid Rad⁺ cells, we used a nonselective assay to identify Ura^- Leu⁺ products of JW3082. Of 343 Ura^- Leu⁺ products analyzed, 10 retained parental structures (intact HO432 sites); these presumably gained a mutation in HO nuclease or in the galactose regulatory network (GALHO^-) and were not analyzed further. Of the remaining 333 products, 319 arose by gene conversion (homozygous at X764), and 14 (4%) arose by imprecise NHEJ (Table 2). Of these, the most common product (6 of 14) had a 2-bp CA insertion, which likely resulted from partial pairing of the 4-base (5'-AACA) overhang followed by filling-in and religation. One product had a single nucleotide insertion within the overhang, and the rest had deletions of 1–17 bp. In all cases, the deletions could be explained as resulting from pairing between microhomologies ranging from 1 to 7 bp. In rad52 mutants, DSB repair at MAT by imprecise NHEJ gives mostly small deletions and insertions, but 28% of deletions were >200 bp in length (KRAMER et al. 1994 ). In JW3082, large deletions extending into LEU2 would give a Ura^- Leu^- phenotype, but among 100 Ura^- Leu^- products examined, none arose by NHEJ; large deletions in JW3082 may be inviable (see DISCUSSION). Ninety-eight were pop-out recombinants; two retained the parental direct repeat structure and may have sustained mutations in LEU2, perhaps as a consequence of DNA polymerase errors during repair synthesis templated from a sister chromatid (STRATHERN et al. 1995 ). Since Ura^- Leu⁺ products comprise 2% of the total (Fig 3 and CHO et al. 1998 ), and 4% of these arise by imprecise NHEJ, 0.1% of DSB repair leading to HO site loss/inactivation involves imprecise NHEJ in HR-competent Rad⁺ haploid yeast.

View larger version (25K): [in this window] [in a new window] [Download PPT slide]   Figure 3. DSB-induced direct repeat recombination. Frequencies of each of the four phenotypic classes, plus totals of all classes, are shown for JW3082 (YKU70) and GJK3465 (yku70). Data represent averages ± SDs for four determinations per strain; 1100–1200 colonies were scored per determination.

yKu70p plays a key role in plasmid NHEJ in yeast (BOULTON and JACKSON 1996B ; MILNE et al. 1996 ). To determine whether imprecise NHEJ of chromosomal DSBs detected in JW3082 was similarly yKu70p dependent, we characterized 127 Ura^- Leu⁺ products from a yku70 derivative of JW3082 (strain GJK3465). Nine products had intact HO432 sites (presumed GALHO^-) and the remainder were long-tract gene conversions. Thus, 0 of 118 DSB repair events in the yku70 mutant involved imprecise NHEJ. This is a significant decrease compared to wild type (P < 0.03; Fisher exact test), confirming that imprecise NHEJ of chromosomal DSBs is yKu70p dependent.

We next sought NHEJ products among DSB-induced Ura^- products in the diploid strain DY3515-13, which carries the same ura3 genes as JW3082 at allelic positions (Fig 1). In this case, products are Ura⁺ (short-tract gene conversion), Ura^- (long-tract gene conversion or imprecise NHEJ), or sectored Ura^+/- (independent G2 events or, less likely, segregation of X764). Of 860 Ura^- products examined, only 2 lacked the NcoI site at position 432 and both had wild-type HO sites (presumed GALHO^-). These 860 Ura^- products represent 1100 products since Ura^- products comprise 80% of the total (Ura⁺ + Ura^-). Thus, imprecise NHEJ in a Rad⁺ diploid comprises <0.1% of total DSB repair.

DSB-induced HR is increased in yku70 mutants:
It was reported that yku70 mutation reduces spontaneous allelic HR by 10- to 40-fold (MAGES et al. 1996 ). We were surprised to find that DSB-induced HR in the haploid yku70 mutant was 1.3-fold higher than the wild-type strain (P < 0.006; Fig 3). The level of gene conversion (Leu⁺ recombinants) was also significantly increased by yku70 mutation (P = 0.05). yku70 also increased HR in an a/a background by 1.5-fold (P < 0.0001; Fig 4). These results can be explained by a model in which yKu70p mediates precise NHEJ of 20–30% of DSBs in wild-type cells that are instead processed by HR in yku70 mutants. Alternatively, yKu70p may directly inhibit HR, although this is unlikely since yku70 did not increase HR in the a/ background (Fig 4). Another possibility is that the increased HR in yku70 reflects increased cleavage by HO nuclease, and we did find that DSB levels were slightly higher in yku70 compared to wild type (Table 3). However, a similar correlation was not seen in the a/a background as yku70 increased HR by 1.5-fold but did not increase DSB levels (Table 3). The slight increase in DSB levels in the haploid yku70 mutant probably reflects reduced DSB repair by precise NHEJ.

View larger version (26K): [in this window] [in a new window] [Download PPT slide]   Figure 4. DSB-induced allelic recombination. Frequencies of Ura+, Ura-, and Ura+/- sectored products plus totals are shown for a/ and a/a strains with wild-type or mutant YKU70. Data represent averages ± SDs for 8–13 determinations per strain, with an average of 1200–1500 colonies scored per determination. NS, not significantly different; *, a statistically significant difference.

yku70 mutation does not reduce mating-type switching in a Rad⁺ background:
yku70 mutants reportedly have reduced levels of GALHO-induced mating-type switching (MAGES et al. 1996 ). We could not assay mating-type switching in our haploid cells because they are MATa-inc. Instead, we assayed GALHO-induced mating-type switching in MATa-inc/MAT diploids. In agreement with our results at ura3, mating-type switching in the yku70 mutant was significantly higher than wild type after a 2-hr induction (P < 0.05); at later times, switching reached similar levels in yku70 and wild-type strains (Fig 5).

View larger version (37K): [in this window] [in a new window] [Download PPT slide]   Figure 5. Mating-type switching in YKU70 and yku70 strains. GALHO-induced mating-type switching was measured in two MATa-inc/MAT strains (DY3515-13, YKU70 and SB3468, yku70) as described in MATERIALS AND METHODS. The average percentage (± SD) of colonies that switched to -maters are shown for four determinations per strain.

MAT heterozygosity enhances HR by yKu-dependent and -independent mechanisms:
The a/a diploid had a total HR frequency significantly lower than that of the a/ diploid (Fig 4). Only a fraction of this difference is yKu70p dependent since even in a yku70 background, HR in the a/a diploid was 2-fold lower than the a/ diploid (P < 0.0001). DSB levels were somewhat lower in a/ than a/a, and this was true in both YKU70 and yku70 backgrounds (Table 4), ruling out the possibility that reduced HR in a/a cells reflects fewer DSBs. Thus, HO-induced HR is reduced in a/a compared to a/ cells, and most of this difference is yKu70p independent, reflecting instead decreased HR in MAT homozygous strains. This decrease in HR closely correlates with decreased cell viability (see DISCUSSION).

View this table: [in this window] [in a new window]   Table 4. DSB and HR levels in a/ and a/a diploids

A single DSB kills 10–20% of MAT homozygous cells, and killing is partially suppressed by yku70 mutation:
We compared cell viability following 6 hr of GALHO expression and repression in the three pairs of matched yKU70 and yku70 haploid (a) and diploid (a/ and a/a) strains. In diploid a/ cells, HO-dependent killing was only 5%, whereas 10–20% killing was observed in a and a/a cells (Fig 6). Interestingly, a/a cells showed significantly less killing in the yku70 mutant compared to wild type (P < 0.05). This trend was also apparent in the haploid and a/ diploid strains, although the differences were smaller and not statistically significant with these sample sizes (P = 0.4 and 0.08, respectively). We conclude that yKu70p has a small negative effect on cell survival following a single DSB in a/a Rad⁺ cells.

View larger version (39K): [in this window] [in a new window] [Download PPT slide]   Figure 6. Cell survival in GALHO-induced cultures. For each cell population, PEs were determined following 6 hr growth in glucose or galactose medium (see MATERIALS AND METHODS). The degree of HO nuclease-dependent cell killing was determined by dividing the galactose PE by the glucose PE for each determination. These ratios were converted to percentages and the averages ± SDs for four to eight determinations per strain were plotted. Values <100% are indicative of HO-dependent cell killing; * indicates a statistically significant difference.

Conversion tract lengths are not affected by yKU70 or MAT status:
The yKu70p/yKu80p heterodimer protects ends from degradation (LEE et al. 1998 ). The longer single-stranded 3' tails in yku70 mutants may influence later steps in HR, such as strand invasion and pairing, and this could enhance hDNA formation and thereby increase gene conversion tract lengths. In our system, gene conversion initiated at HO432 produces Ura⁺ or Ura^- products, with the latter reflecting longer tracts that include X764. Thus, Ura⁺:Ura^- ratios provide an estimate of conversion tract lengths. By this measure, yku70 did not increase tract lengths as Ura^- products comprised 80% of the total in yku70 and YKU70 strains (Fig 4). It is possible that yku70 mutants show extensive 5' end degradation only when HR is disabled (i.e., in rad52 or when no repair template is present; LEE et al. 1998 ).

It has been suggested that MAT heterozygosity enhances HR by enhancing pairing (FRIIS and ROMAN 1968 ; FASULLO and DAVE 1994 ; FASULLO et al. 1999 ; LEE et al. 1999 ), and this might be reflected in increased gene conversion tract lengths. However, we found that 80% of products were Ura^- in both a/ and a/a strains (Fig 4). If MAT heterozygosity enhances HR by enhancing pairing, this is not reflected in increased tract lengths. It is possible that tract lengths are primarily a reflection of branch migration of Holliday junctions and mismatch repair of hDNA, both of which are independent of end-processing and the efficiency of the initial pairing reaction.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Imprecise NHEJ is infrequent in the presence or absence of HR:
In previous studies, the frequency of imprecise NHEJ was estimated by cell survival in rad52 mutants or in the absence of a homologous repair template. Although an early study using an HO swi1 rad52 strain suggested that imprecise NHEJ occurred at a frequency of 1% (WEIFFENBACH and HABER 1981 ), lower frequencies were seen in subsequent studies of rad52 cells suffering DSBs in a dicentric chromosome (0.04% survival) or rad52 MATa cells expressing GALHO (0.01–0.04% survival; KRAMER et al. 1994 ). In Rad⁺ cells lacking a homologous repair template, cell survival reflecting imprecise NHEJ was 0.22% (MOORE and HABER 1996A ). In our study, we used a nonselective assay to estimate the frequency of imprecise NHEJ in Rad⁺ yeast in the presence of a homologous repair template and found a comparable level of imprecise NHEJ (0.1%). Thus, imprecise NHEJ occurs at approximately the same low frequency in the presence or absence of the competing HR pathway.

We found that the rare imprecise NHEJ events in haploid Rad⁺ cells resulted in small 1- to 17-bp deletions and small insertions and confirmed that these arose by a yKu70p-dependent mechanism. KRAMER et al. 1994 also found small insertions and some small deletions in rad52 MATa cells expressing GALHO, but 28% had deletions that ranged from 200 bp to >1 kbp. The formation of large deletions in haploid cells is limited by the proximity of essential genes to the DSB. Large deletions are possible at MAT because MAT is not essential. The closest essential gene to ura3 is TIM9, present only 817 bp downstream of the DSB. In our direct repeat substrate, the 3' end of the LEU2 coding sequence is 950 bp upstream of the DSB. Therefore, symmetric deletions reaching LEU2 would also delete part of TIM9, so it is not surprising that we did not detect large NHEJ-mediated deletions. Imprecise NHEJ was not detected in a/ diploid cells, consistent with the downregulation of NHEJ by MAT heterozygosity (ASTROM et al. 1999 ; LEE et al. 1999 ).

yku70 mutation enhances nuclease-induced HR in Rad⁺ yeast:
There are conflicting reports about yku70 effects on HR and sensitivity to DNA damage. For example, two groups reported that yku70 mutants are hypersensitive to methyl methanesulfonate (MMS) and bleomycin (MAGES et al. 1996 ; MILNE et al. 1996 ), but no effect was seen by a third group for MMS or ionizing radiation (SIEDE et al. 1996 ). MAGES et al. 1996 reported that yku70 reduced spontaneous HR 10- to 40-fold. This result contrasts sharply with the lack of yku70 effect on spontaneous and meiotic HR reported by TSUKAMOTO et al. 1996 and with the enhanced HR in yku70 a and a/a cells that we observed (Fig 3 and Fig 4). MAGES et al. 1996 also reported that yku70 reduced mating-type switching by 3-fold, but we found that yku70 either had no effect or increased mating-type switching (Fig 5); these results may reflect differences in mating-type switching in haploid vs. diploid cells and/or differences in genetic background. MAGES et al. 1996 used W303-derived strains that likely carried a cryptic rad5 mutation (FAN et al. 1996 ; ASTROM et al. 1999 ), whereas our strains, and those used by LEE et al. 1999 that were confirmed to be RAD5, were derived from S288C. Rad5p plays an important role in channeling repair from NHEJ to gene conversion (AHNE et al. 1997 ); however, recent results indicate that the W303 rad5 mutation influences some but not all types of HR (L. SYMINGTON, personal communication). NHEJ assays in RAD5 and rad5 strains have also given conflicting results (AHNE et al. 1997 ; HEGDE and KLEIN 2000 ).

Mating-type control of HR by yKu-dependent and -independent mechanisms:
We assessed repair of a single chromosomal DSB per cell and found that yku70 mutation increased HR by 1.3-fold in haploid yeast and by 1.5-fold in a/a cells, but there was no effect in a/ cells (Fig 4). yku70 mutation increases end processing, resulting in longer 3' single-stranded tails (LEE et al. 1998 ), and these may be better substrates for HR. However, mre11 reduces end processing yet nuclease- induced HR in mre11 occurs at essentially wild-type levels, albeit more slowly (IVANOV et al. 1994 ; TSUBOUCHI and OGAWA 1998 ), suggesting that the extent or rate of end processing does not strongly affect the efficiency of HR. This model also does not account for the lack of yku70 effect on HR in a/ cells. Although NHEJ is downregulated in a/ cells, this is not due to decreased YKU70 expression (GALITSKI et al. 1997 ; ASTROM et al. 1999 ); thus one might expect similar alterations in end processing, and therefore enhanced HR regardless of MAT status, but this was not observed.

We present two alternative models for the enhanced HR in yku70 haploid and a/a strains. The first model is based on the idea that NHEJ and HR compete for repair of DSBs. In this model, yKu70p mediates precise NHEJ of a fraction of HO nuclease-induced chromosomal DSBs in wild-type cells, but these DSBs are processed by HR in yku70 mutants. This interpretation is consistent with the lack of yku70 enhancement of HR (and the lack of imprecise NHEJ) in a/ cells since NHEJ is strongly downregulated in a/ cells (ASTROM et al. 1999 ; LEE et al. 1999 ). Precise NHEJ has been directly detected in assays involving recircularization of linear plasmid DNA transformed into yeast; these events require yKu70p and yKu80p and are detected at lower levels in lif4, lig4, rad50, mre11, and xrs2 mutants, but are RAD52 independent (MEZARD and NICOLAS 1994 ; BOULTON and JACKSON 1996A , BOULTON and JACKSON 1996B , BOULTON and JACKSON 1998 ; HERRMANN et al. 1998 ; LEE et al. 1999 ). Recent studies of EcoRI expression in yeast provided evidence for precise NHEJ of chromosomal DSBs (BARNES and RIO 1997 ; LEWIS et al. 1998 , LEWIS et al. 1999 ). In mammalian cells, nuclease DSBs in transformed plasmid DNA and in chromosomal DNA were shown to be repaired by precise NHEJ (ROTH and WILSON 1985 ; LIN et al. 1999 ). Additional support for the competition model comes from a study of HR in mammalian cells. NHEJ is a major DSB repair pathway in mammalian cells, requiring Ku70, Ku86, and the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs; CRITCHLOW and JACKSON 1998 ). We found that DSB-induced HR was threefold higher in Chinese hamster ovary cells with a defect in DNA-PKcs compared to derivatives carrying a complementing DNA-PKcs cDNA (C. ALLEN, A. KURIMASA, M. BRENNEMAN, D. CHEN and J. A. NICKOLOFF, unpublished results). Thus, elimination of NHEJ has a greater stimulatory effect in mammalian cells than in yeast, consistent with the idea that NHEJ is the dominant repair mode in mammalian cells (LIANG et al. 1998 ). In contrast, wild-type and Ku80-defective hamster cells yielded similar levels of DSB-induced HR (LIANG et al. 1996 ), although this result is questionable because the recombination substrate was present at different chromosomal locations and was therefore subject to position effects (BOLLAG et al. 1989 ; TAGHIAN and NICKOLOFF 1997 ).

The second model suggests that yKu70p interferes with HR. Enhanced end processing in yku70 mutants indicates that yKu70p has an end protection function, but it is important to note that this protection is not limited to the initial end but extends inward as 3' tails are formed (LEE et al. 1998 ). Thus, the presence of yKu at processed ends might interfere with Rad51p function during the formation of nucleoprotein filaments or later during synapsis or strand exchange. In this view, Rad51p function would be enhanced in the absence of yKu70p, and this effect would likely be independent of the length of 3' tails. Note that both models describe yKu-dependent mechanisms by which MAT heterozygosity might regulate HR, either by downregulating NHEJ (competition model) or by influencing yKu70p activity (interference model). The competition and interference models are not mutually exclusive; at present we cannot determine whether only one or both are operative, but our survival data suggest that increased HR in yku70 mutants cannot be explained solely by the absence of precise NHEJ (see below).

Although HR is increased in yku70 a/a cells compared to wild-type a/a cells, HR is still twofold lower than in a/ cells (regardless of YKU70 status; Fig 4). Thus, the reduction of yKu70p-dependent competition or interference in a/ cells does not fully account for the difference in HR levels between a/ and a/a cells, indicating that MAT heterozygosity also enhances HR by a yKu70p-independent mechanism. Our data suggest that most of the difference in HR frequencies between a/a and a/ cells reflects cell killing. In YKU70 strains, the a/ HR frequency was 26%, compared to 8% in a/a cells. The difference of 18% correlates well with the 20% cell killing in a/a cells (note that there is very little killing of a/ strains, regardless of yKU70 status). A similar correlation is apparent in yku70 strains: the a/ HR frequency was 23%, the a/a frequency was 12%, and the difference (11%) was similar to the 9% cell killing in a/a cells. These results suggest that the "missing" recombinants in a/a cells are in fact dead and that HR capacity in a/a cells is insufficient to confer full survival even with only one DSB per cell. In contrast, the higher capacity for HR in a/ cells is sufficient to confer nearly full survival. In this argument, we do not consider the survival value of NHEJ, but focus exclusively on HR. This is because yku70 mutants do not display increased HO-dependent killing compared to wild type (this study and MILNE et al. 1996 ) and Rad⁺ yku70 mutants are not more sensitive to killing by MMS and -rays than wild type (SIEDE et al. 1996 ). Our data indicate that the yKu-independent mechanism by which MAT heterozygosity regulates HR has a stronger effect than the yKu-dependent mechanism(s) (Fig 4).

Slight DSB survival advantage of yku70 mutants:
We found that in an a/a background, yku70 conferred a slight, but significant increase in survival of a single DSB; this trend was also apparent in a and a/ cells (Fig 6). Although MILNE et al. 1996 remarked that a yku70 haploid strain showed wild-type survival following HO-induced cleavage at MAT, survival in yku70 was actually 20% higher than wild type in their experiments. In an mre11 background, yku70 confers sixfold higher survival following exposure to 150 Gy of ionizing radiation (from 1 to 6%; BRESSAN et al. 1999 ). How can inactivation of the yKu70p-dependent DSB repair pathway lead to greater survival of DSB damage? It is doubtful that the observed cell killing reflects chromosome loss because we have previously shown that our diploid cells survive the loss of one copy of chromosome V (NICKOLOFF et al. 1999 ). Yeast cell survival of DSB damage correlates with HR efficiency, as shown here and by others (MORTIMER 1958 ; SAEKI et al. 1980 ; KADYK and HARTWELL 1992 ; FASULLO et al. 1994 ; SCHILD 1995 ). yku70 mutation increased HR in both a and a/a backgrounds, and it is likely that the increased HR underlies increased survival. In a/ cells, yku70 has minimal effect on survival and no effect on HR, which may be a reflection of near-maximum HR levels conferred by MAT heterozygosity. The increase in survival in yku70 mutants cannot be explained solely on the basis of elimination of NHEJ because precise NHEJ produces viable products and imprecise NHEJ is extremely rare. Thus, it appears that yku70-dependent increase in survival is due to elimination of yKu interference with HR. Perhaps a small fraction of DNA ends are blocked from HR by yKu70p, yet fail to engage in a productive NHEJ reaction in a timely fashion, with cell death (or inability to form a colony) perhaps reflecting checkpoint activation. LEE et al. 1998 argued the opposite: that the increased single-stranded DNA in yku70 mutants caused more efficient replication protein A-dependent checkpoint activation. However, in that study there was no possibility for HR because the cells lacked a homologous repair template. Thus, yku70 may increase checkpoint activation only when HR is blocked. It should be possible to gain insight into the roles of checkpoint activation, end processing, and HR in yku70-enhanced cell survival by examining checkpoint mutants, and by using mre11 mutants, which are competent for nuclease-induced HR (IVANOV et al. 1994 ; TSUBOUCHI and OGAWA 1998 ), but display reduced end processing even when combined with yku70 (LEE et al. 1998 ). Also of interest will be studies with lig4 mutants since these have a strong NHEJ defect (BOULTON and JACKSON 1998 ) but are not expected to display altered end processing characteristic of yku70 mutants.

	ACKNOWLEDGMENTS

Helpful comments from Jim Haber, John Petrini, Mark Brenneman, Chris Allen, and Sean Palmer are greatly appreciated. We thank Anna Friedl for providing the yku70 knock-out plasmid pAF1 and Kim Spitz for technical assistance. This research was supported by grant CA55302 from the National Cancer Institute of the National Institutes of Health.

Manuscript received August 7, 2000; Accepted for publication November 13, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AHNE, F., B. JHA, and F. ECKARDT-SCHUPP, 1997  The RAD5 gene product is involved in the avoidance of nonhomologous end-joining of DNA double-strand breaks in the yeast Saccharomyces cerevisiae.. Nucleic Acids Res. 25:743-749[Abstract/Free Full Text].

ASTROM, S. U., S. M. OKAMURA, and J. RINE, 1999  Yeast cell-type regulation of DNA repair. Nature 397:310[Medline].

BARNES, G. and D. RIO, 1997  DNA double-strand break sensitivity, DNA replication, and cell cycle arrest phenotypes of Ku-deficient Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 94:867-872[Abstract/Free Full Text].

BOLLAG, R. J., A. S. WALDMAN, and R. M. LISKAY, 1989  Homologous recombination in mammalian cells. Annu. Rev. Genet. 23:199-225[Medline].

BOULTON, S. J. and S. P. JACKSON, 1996a  Identification of a Saccharomyces cerevisiae Ku80 homologue: roles in DNA double strand break rejoining and in telomeric maintenance. Nucleic Acids Res. 24:4639-4648[Abstract/Free Full Text].

BOULTON, S. J. and S. P. JACKSON, 1996b  Saccharomyces cerevisiae Ku70 potentiates illegitimate DNA double-strand break repair and serves as a barrier to error-prone DNA repair pathways. EMBO J. 15:5093-5103[Medline].

BOULTON, S. J. and S. P. JACKSON, 1998  Components of the Ku-dependent nonhomologous end-joining pathway are involved in telomeric length maintenance and telomeric silencing. EMBO J. 17:1819-1828[Medline].

BRESSAN, D. A., B. K. BAXTER, and J. H. J. PETRINI, 1999  The Mre11-Rad50-Xrs2 protein complex facilitates homologous recombination-based double-strand break repair in Saccharomyces cerevisiae.. Mol. Cell. Biol. 19:7681-7687[Abstract/Free Full Text].

CHO, J. W., G. J. KHALSA, and J. A. NICKOLOFF, 1998  Gene conversion tract directionality is influenced by the chromosome environment. Curr. Genet. 34:269-279[Medline].

CRITCHLOW, S. E. and S. P. JACKSON, 1998  DNA end-joining: from yeast to man. Trends Biochem. Sci. 23:394-398[Medline].

FAN, H. Y., K. K. CHENG, and H. L. KLEIN, 1996  Mutations in the RNA polymerase II transcription machinery suppress the hyperrecombination mutant hpr1 of Saccharomyces cerevisiae.. Genetics 142:749-759[Abstract].

FASULLO, M. and P. DAVE, 1994  Mating type regulates the radiation-associated stimulation of reciprocal translocation events in Saccharomyces cerevisiae.. Mol. Gen. Genet. 243:63-70[Medline].

FASULLO, M., P. DAVE, and R. ROTHSTEIN, 1994  DNA-damaging agents stimulate the formation of directed reciprocal translocations in Saccharomyces cerevisiae. Mutat. Res. 314:121-133[Medline].

FASULLO, M., T. BENNETT, and P. DAVE, 1999  Expression of Saccharomyces cerevisiae MATa and MAT enhances the HO endonuclease-stimulation of chromosomal rearrangements directed by his3 recombinational substrates. Mutat. Res. 433:33-44[Medline].

FISHMAN-LOBELL, J., N. RUDIN, and J. E. HABER, 1992  Two alternative pathways of double-strand break repair that are kinetically separable and independently modulated. Mol. Cell. Biol. 12:1292-1303[Abstract/Free Full Text].

FRIIS, J. and H. ROMAN, 1968  The effect of the mating-type alleles on intragenic recombination in yeast. Genetics 59:33-36[Free Full Text].

GALITSKI, T., A. J. SALDANHA, C. A. STYLES, E. S. LANDER, and G. R. FINK, 1997  Ploidy regulation of gene expression. Science 285:251-254[Abstract/Free Full Text].

HEGDE, V. and H. KLEIN, 2000  Requirement for the SRS2 DNA helicase gene in non-homologous end joining in yeast. Nucleic Acids Res. 28:2779-2783[Abstract/Free Full Text].

HERRMANN, G., T. LINDAHL, and P. SCHAR, 1998  Saccharomyces cerevisiae LIF1: a function involved in DNA double-strand break repair related to mammalian XRCC4. EMBO J. 17:4188-4198[Medline].

HEUDE, M. and F. FABRE, 1993  a/-control of DNA repair in the yeast Saccharomyces cerevisiae: genetic and physiological aspects. Genetics 133:489-498[Abstract].

IVANOV, E. L., N. SUGAWARA, C. I. WHITE, F. FABRE, and J. E. HABER, 1994  Mutations in XRS2 and RAD50 delay but do not prevent mating-type switching in Saccharomyces cerevisiae.. Mol. Cell. Biol. 14:3414-3425[Abstract/Free Full Text].

KADYK, L. C. and L. H. HARTWELL, 1992  Sister chromatids are preferred over homologs as substrates for recombinational repair in Saccharomyces cerevisiae.. Genetics 132:387-402[Abstract].

KRAMER, K. M., J. A. BROCK, K. BLOOM, J. K. MOORE, and J. E. HABER, 1994  Two different types of double-strand breaks in Saccharomyces cerevisiae are repaired by similar RAD52-independent, nonhomologous recombination events. Mol. Cell. Biol. 14:1293-1301[Abstract/Free Full Text].

LEE, S. E., J. K. MOORE, A. HOLMES, K. UMEZU, and R. D. KOLODNER et al., 1998  Saccharomyces Ku70, Mre11/Rad50, and RPA proteins regulate adaptation to G2/M arrest after DNA damage. Cell 94:399-409[Medline].

LEE, S. E., F. PAQUES, J. SYLVAN, and J. E. HABER, 1999  Role of yeast SIR genes and mating type in directing DNA double-strand breaks to homologous and non-homologous repair paths. Curr. Biol. 9:767-770[Medline].

LEWIS, L. K., J. M. KIRCHNER, and M. A. RESNICK, 1998  Requirement for end-joining and checkpoint functions, but not RAD52-mediated recombination, after EcoRI endonuclease cleavage of Saccharomyces cerevisiae DNA. Mol. Cell. Biol. 18:1891-1902[Abstract/Free Full Text].

LEWIS, L. K., J. W. WESTMORELAND, and M. A. RESNICK, 1999  Repair of endonuclease-induced double-strand breaks in Saccharomyces cerevisiae: essential role for genes associated with nonhomologous end-joining. Genetics 152:1513-1529[Abstract/Free Full Text].

LIANG, F., P. J. ROMANIENKO, D. T. WEAVER, P. A. JEGGO, and M. JASIN, 1996  Chromosomal double-strand break repair in Ku80-deficient cells. Proc. Natl. Acad. Sci. USA 93:8929-8933[Abstract/Free Full Text].

LIANG, F., M. G. HAN, P. J. ROMANIENKO, and M. JASIN, 1998  Homology-directed repair is a major double-strand break repair pathway in mammalian cells. Proc. Natl. Acad. Sci. USA 95:5172-5177[Abstract/Free Full Text].

LIN, Y., T. LUKACSOVICH, and A. S. WALDMAN, 1999  Multiple pathways for repair of double-strand breaks in mammalian chromosomes. Mol. Cell. Biol. 19:8353-8360[Abstract/Free Full Text].

MAGES, G. J., H. M. FELDMANN, and E. L. WINNACKER, 1996  Involvement of the Saccharomyces cerevisiae HDF1 gene in DNA double-strand break repair and recombination. J. Biol. Chem. 271:7910-7915[Abstract/Free Full Text].

MANIVASAKAM, P. and R. H. SCHIESTL, 1998  Nonhomologous end joining during restriction enzyme-mediated DNA integration in Saccharomyces cerevisiae.. Mol. Cell. Biol. 18:1736-1745[Abstract/Free Full Text].

MEZARD, C. and A. NICOLAS, 1994  Homologous, homeologous, and illegitimate repair of double-strand breaks during transformation of a wild-type strain and a rad52 mutant strain of Saccharomyces cerevisiae.. Mol. Cell. Biol. 14:1278-1292[Abstract/Free Full Text].

MILNE, G. T., S. JIN, K. B. SHANNON, and D. T. WEAVER, 1996  Mutations in two Ku homologs define a DNA end-joining repair pathway in Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:4189-4198[Abstract].

MOORE, J. K. and J. E. HABER, 1996a  Cell cycle and genetic requirements of two pathways of nonhomologous end-joining repair of double-strand breaks in Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:2164-2173[Abstract].

MOORE, J. K. and J. E. HABER, 1996b  Cell-cycle and genetic requirements of two pathways of nonhomologous end-joining repair of double-strand breaks in Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:2164-2173.

MORTIMER, R. K., 1958  Radiobiological and genetic studies on a polyploid series (haploid to hexaploid) of Saccharomyces cerevisiae.. Radiat. Res. 9:312-326[Medline].

NICKOLOFF, J. A., and M. F. HOEKSTRA, 1998 Double-strand break and recombinational repair in Saccharomyces cerevisiae, pp. 335–362 in DNA Damage and Repair, Vol. 1: DNA Repair in Prokaryotes and Lower Eukaryotes, edited by J. A. NICKOLOFF and M. F. HOEKSTRA. Humana Press, Totowa, NJ.

NICKOLOFF, J. A., J. D. SINGER, M. F. HOEKSTRA, and F. HEFFRON, 1989  Double-strand breaks stimulate alternative mechanisms of recombination repair. J. Mol. Biol. 207:527-541[Medline].

NICKOLOFF, J. A., D. B. SWEETSER, J. A. CLIKEMAN, G. J. KHALSA, and S. L. WHEELER, 1999  Multiple heterologies increase mitotic double-strand break-induced allelic gene conversion tract lengths in yeast. Genetics 153:665-679[Abstract/Free Full Text].

PAQUES, F. and J. E. HABER, 1999  Multiple pathways of recombination induced by double-strand breaks in Saccharomyces cerevisiae.. Microbiol. Mol. Biol. Rev. 63:349-404[Abstract/Free Full Text].

PETES, T. D., R. E. MALONE and L. S. SYMINGTON, 1991 Recombination in yeast, pp. 407–521 in The Molecular and Cellular Biology of the Yeast Saccharomyces: Genome Dynamics, Protein Synthesis, and Energetics, edited by J. R. BROACH, J. R. PRINGLE and E. W. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

RAY, A., I. SIDDIQI, A. L. KOLODKIN, and F. W. STAHL, 1988  Intrachromosomal gene conversion induced by a DNA double-strand break in Saccharomyces cerevisiae.. J. Mol. Biol. 201:247-260[Medline].

ROTH, D. and J. H. WILSON, 1985  Relative rates of homologous and nonhomologous recombination in transfected DNA. Proc. Natl. Acad. Sci. USA 82:3355-3359[Abstract/Free Full Text].

SAEKI, T., I. MACHIDA, and S. NAKAI, 1980  Genetic control of diploid recovery after -irradiation in the yeast Saccharomyces cerevisiae.. Mutat. Res. 73:251-265[Medline].

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHIESTL, R. H. and T. D. PETES, 1991  Integration of DNA fragments by illegitimate recombination in Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 88:7585-7589[Abstract/Free Full Text].

SCHIESTL, R. H., M. DOMINSKA, and T. D. PETES, 1993  Transformation of Saccharomyces cerevisiae with nonhomologous DNA: illegitimate integration of transforming DNA into yeast chromosomes and in vivo ligation of transforming DNA to mitochondrial DNA sequences. Mol. Cell. Biol. 13:2697-2705[Abstract/Free Full Text].

SCHILD, D., 1995  Suppression of a new allele of yeast RAD52 by overexpression of RAD51, mutations in srs2 and ccr4, or mating-type heterozygosity. Genetics 140:115-127[Abstract].

SIEDE, W., A. A. FRIEDL, I. DIANOVA, F. ECKHARDT-SCHUPP, and E. C. FRIEDBERG, 1996  The Saccharomyces cerevisiae Ku autoantigen homologue affects radiosensitivity only in the absence of homologous recombination. Genetics 142:91-102[Abstract].

STRATHERN, J. N., B. K. SHAFER, and C. B. MCGILL, 1995  DNA synthesis errors associated with double-strand break repair. Genetics 140:965-972[Abstract].

SWEETSER, D. B., H. HOUGH, J. F. WHELDEN, M. ARBUCKLE, and J. A. NICKOLOFF, 1994  Fine-resolution mapping of spontaneous and double-strand break-induced gene conversion tracts in Saccharomyces cerevisiae reveals reversible mitotic conversion polarity. Mol. Cell. Biol. 14:3863-3875[Abstract/Free Full Text].

TAGHIAN, D. G., and J. A. NICKOLOFF, 1996 Subcloning strategies and protocols, pp. 221–235 in Basic DNA and RNA Protocols, edited by A. HARWOOD. Humana Press, Totowa, NJ.

TAGHIAN, D. G. and J. A. NICKOLOFF, 1997  Chromosomal double-strand breaks induce gene conversion at high frequency in mammalian cells. Mol. Cell. Biol. 17:6386-6393[Abstract].

TSUBOUCHI, H. and H. OGAWA, 1998  A novel mre11 mutation impairs processing of double-strand breaks of DNA during both mitosis and meiosis. Mol. Cell. Biol. 18:260-268[Abstract/Free Full Text].

TSUKAMOTO, Y., J.-I. KATO, and H. IKEDA, 1996  Hdf1, a yeast Ku-protein homologue, is involved in illegitimate recombination but not in homologous recombination. Nucleic Acids Res. 24:2067-2072[Abstract/Free Full Text].

WEIFFENBACH, B. and J. E. HABER, 1981  Homothallic mating type switching generates lethal chromosome breaks in rad52 strains of Saccharomyces cerevisiae.. Mol. Cell. Biol. 1:522-534[Abstract/Free Full Text].

WENG, Y.-S. and J. A. NICKOLOFF, 1998  Evidence for independent mismatch repair processing on opposite sides of a double-strand break in