The Chromosome Bias of Misincorporations During Double-Strand Break Repair Is Not Altered in Mismatch Repair–Defective Strains of Saccharomyces cerevisiae

Corresponding author: Jeffrey N. Strathern, Gene Regulation and Chromosome Biology Laboratory, NCI-Frederick Cancer Research and Development Center, ABL-Basic Research Program, Box B, Bldg. 539, Frederick, MD 21702-1201, strather{at}ncifcrf.gov (E-mail).

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Recombinational repair of a site-specific, double-strand DNA break (DSB) results in increased reversion frequency for nearby mutations. Although some models for DSB repair predict that newly synthesized DNA will be inherited equally by both the originally broken chromosome and the chromosome that served as a template, the DNA synthesis errors are almost exclusively found on the chromosome that had the original DSB (introduced by the HO endonuclease). To determine whether mismatch repair acts on the template chromosome in a directed fashion to restore mismatches to the initial sequence, these experiments were repeated in mismatch repair-defective (pms1, mlh1, and msh2) backgrounds. The results suggest that mismatch repair is not responsible for the observed bias.

SEVERAL years ago, DRAKE 1994 reviewed the first 75 years of research on the origin of mutations and made predictions of what would be discovered in the early twenty-first century. While the progress in basic research in this field has been impressive, and the pragmatic consequences of defects in processes that limit replication fidelity are increasingly apparent, he highlighted the multiple areas influencing the origin of spontaneous and induced mutations that are not yet understood. It is encouraging to see that the progress he predicts will be made on the multiple issues that are still unresolved. Thus far, two main pathways for the origin of base substitution mutations have been established: alterations of the template (including those caused by chemical modification and irradiation) resulting in defective coding capacity, and misincorporations on undamaged templates during replicative synthesis. Chemical alterations of the template are often viewed as premutagenic lesions because they are subject to various repair pathways that can remove or reverse such lesions (FRIEDBERG 1988 ). Similarly, the initial misincorporations by DNA polymerase should be considered premutagenic. Elegant studies led to the view that net base substitution rates reflect the combined consequences of initial misincorporation, followed by editing via 3'–5' exonuclease activity of the DNA polymerases and the directed mismatch repair processes (KUNKEL and LOEB 1981 ; KORNBERG and BAKER 1991 ; BECKMAN and LOEB 1993 ; SCHAAPER 1993 ). We are studying the relative effects of these processes on the fidelity of DNA synthesis during recombination.

The repair of double-strand DNA breaks (DSB) in yeast occurs predominantly by recombinational mechanisms that involve DNA synthesis. This recombination-associated repair synthesis provides an additional opportunity for replication errors. Indeed, repair of a site-specific DSB induced by HO endonuclease, which cuts only once in the yeast genome, causes a several 100-fold increase in the reversion rate of a nearby (0.3-kb) marker (STRATHERN et al. 1995 ). It has been demonstrated that the mutation rate is elevated in meiosis (MAGNI and VON BORSTEL 1962 ) and that such mutations are positively correlated with nearby crossover events (MAGNI 1964 ). A correlation of spontaneous mutation and mitotic recombination leading to crossovers has also been demonstrated (ESPOSITO and BRUSCHI 1993 ). Combined, these observations suggest that elevated DNA synthesis error rates may be a general feature of recombination processes. Perhaps the elevated mutation rate associated with recombination contributes significantly to the overall mutation rate. We demonstrated that the DNA synthesis associated with DSB repair has a relatively high mutation rate (STRATHERN et al. 1995 ). Our current work focuses on the relative contributions of polymerase errors and escape from directed mismatch repair to these recombination-associated errors. The translesion polymerase encoded by REV3 (NELSON et al. 1996 ) is responsible for most of the base substitution errors during DSB repair (HOLBECK and STRATHERN 1997 ). In this report, we investigate the contribution of mismatch repair to the rate and distribution of DNA misincorporations during DSB repair.

In the experiments demonstrating that there is an elevated mutation rate associated with DSB-initiated mitotic recombination, we noted that the mutations are nonrandomly distributed between the two interacting chromosomes (STRATHERN et al. 1995 ). Because only one chromosome had the recognition site for the endonuclease (HO) that made the DSB (Figure 1A), the chromosome that had the DSB can be identified in cases where exchange of outside markers does not occur. The DSB repair model, as detailed by SZOSTAK et al. 1983 , predicts that newly synthesized DNA, and hence new errors, are inherited on both the originally broken chromosome and the chromosome that served as the template (Figure 1). In contrast, we found that the errors are inherited almost exclusively by the chromosome that originally had the DSB. We entertained two classes of models for this bias. In one class of models, the noncrossover events are the result of mechanisms in which the newly synthesized DNA, and hence all the errors, are inherited by the cut chromosome. All the newly synthesized DNA is inherited by the repaired chromosome if a double Holliday junction intermediate is resolved by a topoisomerase (Figure 2B; NASMYTH 1982 ; THALER et al. 1987 ; HASTINGS 1988 ; MCGILL et al. 1989 ) rather than by strand cleavage or if replication proceeds by a conservative mechanism (Figure 2C).

View larger version (42K): In this window In a new window Download PPT slide   Figure 1. —DSB repair model applied to the reversion of trp1-488. (A) Physical description of the TRP1-HIS3 interval. The TRP1 and HIS3 genes were inserted into an EcoRI site centromere proximal to MAT. Mutations in the HIS3 gene were made by filling in the NdeI site at codon 64 (his3-192) or by filling in the AspI site at codon 207 (his3-621). Both MAT alleles (MAT-inc and MATa-inc) are resistant to cutting by HO endonuclease. The HO site was inserted into a polylinker between trp1 and his3 on the cry1 MATa-inc chromosome. Both chromosomes carry the trp1-488 mutation. The trp1-488 allele was made by site-directed oligonucleotide mutagenesis to change two bases and create an SpeI site (from ACTGGG to ACTAGT) with an in-frame stop codon at codon 163 (trp1-488). (B) The DNA strands of the broken chromosome are white and are subject to recision. The uncut chromosome used as a template for repair is grey. (C) Newly synthesized DNA, copied from the uncut chromosome, is crosshatched. The lengths of the double-strand gap, the half gaps, and the regions of symmetric strand exchange can be variable. (D and E) Resolution without crossing over by cleavage of Holliday junctions results in both the cut chromosome and the repair template chromosome receiving newly synthesized DNA (and, hence, the potential for errors). Note the potential positioning of strand discontinuities.

View larger version (25K): In this window In a new window Download PPT slide   Figure 2. —Resolution models that restrict revertants to the repaired chromosome. (A) Asymmetric gap repair followed by resolution of the two Holliday junctions with the sense confining the new synthesis to the repaired chromosome. (B) Resolution of a symmetric double Holliday junction intermediate by topoisomerase. (C) Traveling replication bubble yielding conservative DNA replication.

An alternative model for the bias in inheritance of errors is that resolution proceeds by cleavage of Holliday junctions. This allows intermediates carrying errors on either chromosome, but the correction of mismatches on the template chromosome is biased toward restoration, while correction of the mismatches on the repaired chromosome is random. We postulated that if mismatch correction were directed by nicks, a different distribution of nicks on the two chromosomes would promote the observed bias (STRATHERN et al. 1995 ). Specifically, we noted that resolution by cleavage of the two Holliday junctions to produce a noncrossover DSB repair results in the initially uncut (template) chromosome having one original strand with no nicks and a second strand carrying any newly synthesized DNA that would include strand discontinuities (Figure 1D and Figure E). The presence of the nicks on the strand with the newly synthesized DNA (and hence any new misincorporations) might be signals for the directed repair of mismatches, as has been demonstrated for Escherichia coli mutS system (LANGLE-ROUAULT et al. 1987 ; LAHUE et al. 1989 ; MODRICH 1991 ) and for eukaryotes (HOLMES et al. 1990 : THOMAS et al. 1991). This would be true for the template chromosome for either pairing of cleavages of the two Holliday junctions that would lead to a noncrossover (Figure 1D vs. E). In contrast, the DNA of the chromosome that originally had the DSB would have nicks on both strands and hence not be repaired in a directed fashion. This proposal shares some features with that made by ALANI et al. 1994 , and elaborated upon by SCHWACHA and KLECKNER 1995 , for how the distribution of nicks associated with the resolution of meiotic recombination intermediates could bias the repair of heteroduplex DNA and help explain the paucity of certain classes of recombinants. In the experiments presented here, we test the hypothesis that mismatch correction is responsible for restoration of mismatches on the template chromosome by monitoring the reversion of a marker near the site of a DSB in strains that are defective in mismatch repair because of mutations in the PMS1, MLH1 or MSH2 genes.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains:
The yeast strains used are listed in Table 1. The basic TRP1-HIS3 module and the positions of the alleles are given in Figure 1 and described in MCGILL et al. 1990 , MCGILL et al. 1993 and STRATHERN et al. 1995 . The pms1::LEU2 disruption allele was made by single-step gene disruption using a plasmid (pWJ401) provided by Dr. RODNEY ROTHSTEIN. The mlh1-267::LEU2 allele was made by transformation with a fragment of pJS267, which has the LEU2 gene as a substitution for codons 30–702 of MLH1 flanked by 679 bases 5' and 476 bases 3' of the deleted MLH1 region. The msh2-265::LEU2 allele was made by transformation with a fragment of pJS265 that has the LEU2 gene as a substitution for codons 11–768 of MSH2 flanked by 608 bases 5' and 536 bases 3' of the deleted MSH2 region.

Genetic analysis:
Coupling of the TRP1 revertant alleles to MAT and CRY1 was established by classical genetics, as described in MCGILL et al. 1990 . In brief, patches of the a/ diploid revertants were sporulated. Coupling of the cry1 and MAT alleles was determined by replica plating the sporulated cultures onto YEPD plates containing 1 mg/ml cryptopleurine. The mating type of the resulting patch of cryptopleurine-resistant spores was then determined by the ability to mate to MATa and MAT strains (DC14 and DC17). The coupling of the revertant TRP1 allele to MAT was determined from the sporulated cultures by the ability to mate to and complement the Trp^- defect of MATa and MAT trp1 tester strains (GRY7 70 and GRY7 72). In the fluctuation tests, the extent of induction of the HO endonuclease–mediated DSB was determined by testing several hundred Ura⁺ colonies for changes in the his3 and MAT alleles and for loss of the HO site. Specifically, the diploids were tested for mating phenotype to determine whether they retained the a/ mating phenotype that is indicative of heterozygosity of MATa and MAT, and they were tested for heterozygosity of the his3 alleles by scoring their ability to give rise to UV-induced His⁺ recombinants. Those diploids that were heterozygous for the his3 alleles and carried the pGALHO plasmid were tested for the presence of the HO cleavage site by monitoring the ability of growth on galactose to promote the formation of His⁺ recombinants. The revertants in the experiment that were used to demonstrate the chromosome bias came from independent cultures grown either in liquid or as patches started from single colonies. The galactose-induced cultures included some revertants that were spontaneous in origin. These could be identified as strains that were heterozygous for the his3 alleles and that retained the HO site and, hence, the ability to be galactose induced to HIS3.

Media:
The media in these experiments were prepared as described in SHERMAN et al. 1986 and MCGILL et al. 1990 . Galactose induction was performed by shifting cells from synthetic complete medium minus uracil plus 5% raffinose (SC-Ura+raffinose) to SC-Ura plus 2% galactose overnight. Aliquots were titered and plated to detect His⁺ recombinants and Trp⁺ revertants. Frequencies were determined from the median value of 11 independent cultures.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Reversion of mutant trp1 alleles near the site of a DSB repair event is elevated only on the chromosome that had experienced the DSB (STRATHERN et al. 1995 ). We can induce a site- and chromosome-specific DSB in these diploid strains because there is a recognition site for the HO endonuclease between the trp1 and his3 genes on the cry1 MATa-inc chromosome. The MATa-inc and MAT-inc alleles carry defective HO recognition sites, thus, the site between trp1 and his3 is the only available target for HO. The HO gene is under the control of a galactose-regulated promoter ( JENSEN and HERSKOWITZ 1984 ). If the observed bias favoring the inheritance of the reverted TRP1 allele by the cut chromosome reflects directed mismatch repair of errors made on the template chromosome, then we predict that removal of the mismatch repair system should allow recovery of TRP1 revertants made during DSB repair on the template chromosome.

Reversion of trp1- 488 in a pms1 background:
We tested this hypothesis first in diploids disrupted for PMS1 (KRAMER et al. 1989 ), one of the yeast homologs of the E. coli mutL gene. The mutL protein appears to be involved in defining the strandedness of mismatch repair by recognizing nicks made by mutH in the unmethylated strand of newly replicated DNA (MODRICH and LAHUE 1996 ). We monitored reversion of the trp1-488 allele in a diploid (GRY1657) homozygous for the pms1::LEU2 disruption, and we observed a 10-fold higher spontaneous reversion frequency consistent with the expectations of a mutator phenotype for pms1 mutants (Table 2). When this strain was grown on galactose to induce HO endonuclease and initiate DSB repair, we observed a further increase in the reversion frequency of the trp1-488 allele, but not to a level demonstrably different than what was observed for the Pms⁺ strain. The pms1 mutation did not alter the observed efficiency of induction of the DSB or the induction of HIS3 recombinants on the other side of the DSB.

Meiotic analysis was used to determine which chromosome carried the reverted TRP1 allele and whether the reversion was accompanied by an exchange of the outside markers CRY1 and MAT (MCGILL et al. 1990 ). For noncrossover diploids, the chromosome that had the DSB was identified by the cry1 and MATa-inc alleles, while the template chromosome was CRY1 MAT-inc. Nearly all the spontaneous revertants occurred without an associated crossover, and the revertant TRP1 allele was found equally often on the two chromosomes (Table 3). In the HO-stimulated revertants, 23% of the events occurred without crossover. When HO endonuclease was expressed, the TRP1 allele was preferentially found on the cut chromosome (cry1 MATa-inc) in the pms1-defective strain similar to the wild-type diploid. Thus, the chromosome bias in the recovery of the revertant allele was not PMS1 dependent.

Reversion of trp1- 488 in a mlh1 background:
The MLH1 gene is another yeast homolog of the mutL gene of E. coli. The diploid strain GRY1731, which is homozygous for the mlh1::LEU2 disruption, showed a spontaneous reversion frequency for trp1-488 sixfold higher than wild-type, consistent with the expectations of a mutator phenotype for mlh1 mutants (Table 2). When HO endonuclease was induced in the mlh1 diploid, we observed a further increase in the frequency of Trp⁺ revertants to a level similar to the induced level in the wild-type strain. The mlh1 mutation did not affect the induction of HIS3 recombinants. As observed for pms1, there was no chromosome bias for spontaneous revertants, but we observed a strong bias in DSB-associated revertants favoring the recovery of the revertant allele on the cut chromosome (cry1 MATa-inc), as defined in the 19% of revertants not associated with a crossover. Thus, MLH1 is not required for the asymmetry in the genetic coupling of the revertant allele.

Reversion of trp1-488 in a msh2 background:
We monitored the reversion of the trp1-488 allele in a diploid (GRY1732) defective in the MSH2 gene. The msh2 diploid exhibited a spontaneous mutator phenotype, but among cells with an HO-induced DSB, the reversion frequency was similar to that observed in wild-type cells and the pms1 and mlh1 strains described above (Table 2). The spontaneous revertants isolated from msh2 cultures were randomly distributed between the two chromosomes (Table 3). In contrast, the DSB-associated revertants showed the same biased distribution, favoring the cut chromosome over the template chromosome seen in wild-type or the pms1 and mlh1 mutant backgrounds.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In addition to its role in genome duplication, DNA synthesis is a crucial part of some mechanisms of DNA damage repair and recombination. The net stability of the sequence of the genome depends on the fidelity of each of these processes. The elevated mutation rates in cells that are defective in mismatch correction demonstrate the importance of this process in mutation avoidance (reviewed in MODRICH and LAHUE 1996 ). However, the relative importance of mismatch correction in the various classes of DNA synthesis remains unclear. Understanding the origin of spontaneous mutations will require the determination of what DNA polymerases and accessory factors contribute to these multiple DNA synthesis processes. In these experiments, we monitored the fidelity of DNA synthesis associated with DSB repair. Previous experiments demonstrated that there is an elevated mutation rate in the region (0.3 kb) of DSB repair events, and that those mutations are recovered far more often on the repaired chromosome than on the template chromosome (STRATHERN et al. 1995 ). In this report, we focused on the chromosome bias in the recovery of misincorporations associated with DSB repair.

Just as the distribution of gene convertants and postmeiotic segregants and their association with meiotic recombination constrains models to explain their origins (HOLLIDAY 1964 ; MESELSON and RADDING 1975 ), the asymmetric inheritance of mutations associated with DSB repair constrains models for how they are generated. For the purposes of this discussion, the excess of 5:3 tetrads over Ab5:3 tetrads in yeast provides an important parallel (FOGEL et al. 1981 ; RADDING 1982 ). In 5:3 tetrads, the sectored colony carries the chromosome defined as the recipient, whereas in Ab5:3 tetrads, the sectored colony carries the chromosome defined as the donor. This distinction can only be made for tetrads that do not have a crossover for outside markers. As discussed below, the mechanisms that constrain against Ab5:3 in meiosis can also function as mechanisms that determine the chromosome that inherits errors associated with mitotic DSB repair.

The recombination events studied here are initiated by site-specific, double-strand cleavage. The symmetric form of the DSB repair model for genetic recombination as described by SZOSTAK et al. 1983 allows the generation of 6:2 gene convertant tetrads resulting from either the repair of double-strand gaps or by the mismatch correction of heteroduplexes in regions of single- or symmetric-strand exchange. Repair of a double-strand gap requires that both strands of DNA spanning the gap be resynthesized, while repair of a single-strand gap requires only one strand of DNA to be synthesized. In the symmetric form of the DSB repair model, newly synthesized DNA is inherited by both the repaired and the template chromosome (Figure 1). In the experiments reported here, we tested the hypothesis that the failure to recover revertants on the template chromosome results from directed mismatch repair of base substitutions leading to restoration on the template chromosome. We can define the template chromosome in repair events that are not associated with crossing over because the recognition site for HO endonuclease is present on only one of the two chromosomes. This is in contrast to most meiotic experiments, where the two alleles have similar probabilities of acting as initiator and the donor is inferred from the resulting postmeiotic distribution of markers.

Resolution of a double Holliday junction joint molecule to produce a noncrossover involves cleavage of the same strand of each duplex at both Holliday junctions (SZOSTAK et al. 1983 ; ALANI et al. 1994 ). On the template chromosome, the newly synthesized DNA (and hence any errors) will always be on the strand that was cut to resolve the Holliday junctions (Figure 1D and Figure E). The use of those strand discontinuities to direct correction of misincorporations made during DSB repair would result in restoration of the template chromosome sequence (STRATHERN et al. 1995 ). On the repaired chromosome, the newly synthesized DNA will always be on the strand that was not cleaved to resolve the Holliday junctions. The use of those strand discontinuities to direct mismatch repair would result in fixation of the new sequence on the repaired chromosome. SCHWACHA and KLECKNER 1995 proposed a similar model for how the nicks left from the resolution of Holliday junctions could account for directed mismatch repair and the paucity of Ab5:3 tetrads. We also noted that the strand discontinuities at the end of the gap repair synthesis might also still be present to guide mismatch correction. On the template chromosome, the strand break at the end of synthesis would be on the same strand as the strand breaks, resulting from resolution of the Holliday junctions. Nick-directed mismatch repair would then promote restoration. On the repaired chromosome, the strand break at the end of the synthesis would be on the opposite strand from the strand break from the cleavage of the Holliday junctions. The presence of breaks on both strands could lead to randomized mismatch correction.

Our results are not consistent with the model that the chromosome bias in recovering revertants associated with DSB repair is the result of nick-directed mismatch correction. The protein encoded by the mutS homolog MSH2 (REENAN and KOLODNER 1992 ) is directly involved in mismatch recognition (ALANI et al. 1995 ; ALANI 1996 ) and, hence, is a prime candidate for an essential function in any model invoking directed mismatch repair. The proteins encoded by the mutL homologs MLH1 and PMS1 form a heterodimer, and they are both required for mismatch repair (PROLLA et al. 1994A , PROLLA et al. 1994B ). We tested both pms1 and mlh1 strains because it is possible that they provide different functions and could act independently in correcting mistakes made during DNA synthesis associated with DSB repair. We found, however, that defects in PMS1, MLH1 or MSH2 do not change the bias favoring recovery of the revertants on the repaired rather than the template chromosome. Furthermore, the frequency of revertants associated with DSB repair is the same in the mismatch repair-defective strains as it is in the wild-type background. We do not see the twofold increase in mutation frequency in the repair-defective strains that was predicted from the model that mismatches are repaired randomly in wild-type strains. This suggests that base substitutions made during DSB repair were not substrates for the products of these genes, even when the misincorporation was on the repaired chromosome. It remains possible that there are mismatch detection mechanisms independent of PMS1, MLH1, and MSH2 that are dedicated to the DSB repair process. Such proteins could be specifically associated with the polymerases involved in DSB repair and/or those proteins specifically involved in the formation or resolution of the recombination intermediate.

The elevated reversion of the trp1-488 nonsense allele near the site of a DSB repair is almost completely dependent on REV3, suggesting that the translesion polymerase (Pol ) encoded by REV3 (NELSON et al. 1996 ) is recruited to recombination intermediates and makes misincorporation errors on these presumably undamaged templates (HOLBECK and STRATHERN 1997 ). Thus, the revertants of trp1-488 studied here could be the result of a polymerization step that is not reflective of the majority of DSB repair syntheses. That is, they might represent events in which the translesion polymerase is recruited because the template is not used efficiently by the standard polymerases. For that reason, they might reflect a pathway that does not have associated mismatch repair functions required for replication fidelity. In its role as a translesion polymerase, it might be counterproductive for Pol to be coupled to a mismatch detection system that is biased toward removing the newly synthesized strand. One would still be left with the puzzle of why lesions made by Pol are preferentially inherited by the repaired chromosome.

TRAN et al. 1996 found that the ability to repair single base deletions is dependent on how they are made. They observed that the rate of single base frameshift reversions (deletions) are elevated in a rad52 background, but this mutator phenotype of rad52 is not made more dramatic by an additional defect in mismatch repair (pms1). They conclude that the deletions are not made during semiconservative DNA synthesis, but instead, that they occur during an error-prone process that does not yield a substrate that can be detected by mismatch repair proteins. ROCHE et al. (1995) demonstrated that the mutator phenotype caused by rad52 is dependent on REV3 (NELSON et al. 1996 ). Combined, these results also suggest that mutations made by Pol are not substrates for mismatch repair.

Three other classes of models have been proposed for how the double Holliday junction might be a common intermediate in meiotic recombination and not yield Ab5:3 tetrads. SZOSTAK et al. 1983 recognized that the symmetric intermediate predicted the common formation of Ab5:3 tetrads, and they proposed constraints on the formation of regions of strand exchange and the resolution of the Holliday junctions that would prohibit those classes of tetrads. Specifically, they suggested that the extent of asymmetric strand exchange was much greater on one side of the "bubble," and that resolution of the Holliday junctions was controlled so that the exchanged strand was inherited by the chromosome that had the initiating DSB. Note that in this view, the template chromosome gets most of the new DNA synthesis. A related model can be used to explain the distribution of revertants associated with DSB repair (Figure 2A). In this view, the region of double-strand gap would have to be small relative to the single-strand gap, and the single-strand recision (and hence required resynthesis) would have to be primarily in one direction from the DSB. In contrast to the model for meiotic recombination, the resolution of the Holliday junctions in mitotic DSB repair would have to be biased so that the repaired chromosome received the strand with the newly synthesized DNA.

A second way to resolve a symmetric double Holliday junction as a noncrossover without leaving heteroduplexes on both chromosomes is to use topoisomerases rather than cleavage of the Holliday junctions (NASMYTH 1982 ; THALER et al. 1987 ; HASTINGS 1988 ; MCGILL et al. 1989 ). Resolution of a double Holliday junction by topoisomerases results in all the newly synthesized DNA being recovered on the repaired chromosome (Figure 2B). Support for this mechanism of resolution has been reported, based on the strandedness of heteroduplexes on opposite sites of an initiation site in meiosis (GILBERTSON and STAHL 1996 ) and in homothallic switching (MCGILL et al. 1989 ).

DSB repair can be accomplished without both sides of the break invading the template chromosome (RESNICK 1976 ). In this class of models, DNA synthesis proceeds on the template chromosome from only one side until the region of the DSB has been spanned (Figure 2C). Interaction of the newly synthesized DNA and the other side of the broken chromosome can allow repair of the DSB without the formation of a double Holliday junction intermediate. This pathway does not produce reciprocal crossovers, but it can yield the noncrossover events that are the focus of this study.

The results presented here demonstrate that not all DNA synthesis is equivalent with regard to mutation avoidance mechanisms. This was first demonstrated by SANTOS and DRAKE 1994 in experiments showing that the net misincorporation rate for T4 replication is higher than for its host, E. coli, and is not subject to mismatch correction. Starvation in E. coli leads to elevated mutation rates (reviewed by FOSTER 1993 ), which may reflect decreased mismatch repair capacity (HARRIS et al. 1997 ), suggesting that cells may be able to adjust the fidelity of DNA synthesis in response to the environment. Localized elevated mutation rates have been implicated in the origin of the somatic diversity of immunoglobin genes (NEUBERGER and MILSTEIN 1995 ), but it remains to be determined whether that process is subject to mismatch correction.

Our results highlight the complexity of the issue of the origin of spontaneous mutations. Mismatch correction is a critical component of the processes that define the fidelity of genome replication. Defects in MSH2, PMS1 or MLH1 cause substantial increases in the levels of spontaneous mutagenesis in yeast. These functions, however, appear to play little role in defining the mutation frequency associated with DSB repair. Furthermore, while these functions have been reported to alter mitotic recombination (DATTA et al. 1997 ; NEGRITTO et al. 1997 ), as well as the pattern of inheritance of gene conversions and PMS events in meiosis (BORTS et al. 1990 ; HUNTER and BORTS 1997 ), they did not alter the chromosome bias in the recovery of revertants of trp1-488 stimulated by DSB repair. It remains to be determined whether that bias represents the use of alternate mismatch detection proteins or whether it is a consequence of the DSB repair strand mechanics.

	ACKNOWLEDGMENTS

We thank ALISON RATTRAY for helpful comments and JOAN HOPKINS for manuscript preparation. Research was sponsored by the National Cancer Institute, Department of Health and Human Services (DHHS), under contract with ABL. The contents of this publication do not necessarily reflect the views or policies of the DHHS, nor does any mention of trade names, commercial products or organizations imply endorsement by the U.S. Government.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALANI, E. R., R. A. G. REENAN, and R. D. KOLODNER, 1994  Interaction between mismatch repair and genetic recombination in Saccharomyces cerevisiae.. Genetics 137:19-39[Abstract].

ALANI, E., 1996  The Saccharomyces cerevisiae Msh2 and Msh6 proteins form a complex that specifically binds to duplex oligonucleotides containing mismatched DNA base pairs. Mol. Cell. Biol. 16:5604-5615[Abstract].

ALANI, E., N. W. CHI, and R. KOLODNER, 1995  The Saccharomyces cerevisiae Msh2 protein specifically binds to duplex oligonucleotides containing mismatched DNA base pairs and insertions. Genes Dev. 9:234-247[Abstract/Free Full Text].

BECKMAN, R. A. and L. A. LOEB, 1993  Multi-stage proofreading in DNA replication. Q. Rev. Bio. 26:225-231.

BORTS, R. H., W. Y. LEUNG, W. KRAMER, B. KRAMER, and M. WILLIAMSON et al., 1990  Mismatch repair-induced meiotic recombination requires the pms1 gene product. Genetics 124:573-584[Abstract].

DATTA, A., M. HENDRIX, M. LIPSITCH, and S. JINKS-ROBERTSON, 1997  Dual roles for DNA sequence identity and the mismatch repair system in the regulation of mitotic crossing-over in yeast. Proc. Natl. Acad. Sci. USA 94:9757-9762[Abstract/Free Full Text].

DRAKE, J. W., 1994  Looking backward on a century of mutation research. Environ. Mol. Mutagen. 23(Suppl.):11-14.

ESPOSITO, M. S. and C. V. BRUSCHI, 1993  Diploid yeast cells yield homozygous spontaneous mutations. Curr. Genet. 23:430-434[Medline].

FOGEL, S., R. MORTIMER and K. LUSNAK, 1981 Mechanisms of meiotic gene conversion, or "wanderings on a foreign strand," pp. 289–339 in The Molecular Biology of the Yeast Saccharomyces: Life Cycle and Inheritance, edited by J. N. STRATHERN, E. W. JONES and J. R. BROACH. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

FOSTER, P. L., 1993  Adaptive mutation: the uses of adversity. Annu. Rev. Microbiol. 47:467-504[Medline].

FRIEDBERG, E. C., 1988  Deoxyribonucleic acid repair in the yeast Saccharomyces cerevisiae.. Microbiol. Rev. 52:70-102[Free Full Text].

GILBERTSON, L. A. and F. W. STAHL, 1996  A test of the double-strand break repair model for meiotic recombination in Saccharomyces cerevisiae.. Genetics 144:27-41[Abstract].

HARRIS, R. S., G. FENG, K. J. ROSS, R. SIDHU, and C. THULIN et al., 1997  Mismatch repair protein MutL becomes limiting during stationary-phase mutation. Genes Dev. 11:2426-2437[Abstract/Free Full Text].

HASTINGS, P. J., 1988  Recombination in the eukaryotic nucleus. BioEssays 9:61-64[Medline].

HOLBECK, S. L. and J. N. STRATHERN, 1997  A role for REV3 in mutagenesis during double-strand break repair in Saccharomyces cerevisiae.. Genetics 147:1017-1024[Abstract].

HOLLIDAY, R., 1964  A mechanism for gene conversion in fungi. Genet Res. 5:282-304.

HOLMES, J. J., S. CLARK, and P. MODRICH, 1990  Strand specific mismatch correction in nuclear extracts of human and Drosophila melanogaster cell lines. Proc. Natl. Acad. Sci. USA 87:5837-5841[Abstract/Free Full Text].

HUNTER, N. and R. H. BORTS, 1997  Mlh1 is unique among mismatch repair proteins in its ability to promote crossing-over during meiosis. Genes Dev. 11:1573-1582[Abstract/Free Full Text].

JENSEN, R. and I. HERSKOWITZ, 1984  Directionality and regulation of cassette substitution in yeast. Cold Spring Harbor Symp. Quant. Biol. 49:97-104[Abstract/Free Full Text].

KRAMER, W., B. KRAMER, M. S. WILLIAMSON, and S. FOGEL, 1989  Cloning and nucleotide sequence of DNA mismatch repair gene PMS1 from Saccharomyces cerevisiae: homology to procaryotic MutL and HexB.. J. Bacteriol. 171:5339-5346[Abstract/Free Full Text].

KORNBERG, A., and T. A. BAKER, 1991 DNA Replication. W. H. Freeman and Co., Oxford.

KUNKEL, T. A. and L. A. LOEB, 1981  Fidelity of mammalian DNA polymerases. Science 213:765-767[Abstract/Free Full Text].

LAHUE, R. S., K. G. AU, and P. MODRICH, 1989  DNA mismatch correction in a defined system. Science 245:160-164[Abstract/Free Full Text].

LÄNGLE-ROUAULT, F., G. MAENHAUT-MICHEL, and M. RADMAN, 1987  GATC sequences, DNA nicks and the MutH function in Escherichia coli mismatch repair. EMBO J. 6:1121-1127[Medline].

MAGNI, G. E., 1964  Origin and nature of spontaneous mutations in meiotic organisms. J. Cell. Comp. Physiol. 64:165-172.

MAGNI, G. E. and R. C. VON BORSTEL, 1962  Different rates of spontaneous mutation during mitosis and meiosis in yeast. Genetics 47:1097-1108[Free Full Text].

MCGILL, C. B., B. K. SHAFER, and J. N. STRATHERN, 1989  Coconversion of flanking sequences with homothallic switching. Cell 57:459-467[Medline].

MCGILL, C. B., B. K. SHAFER, D. R. HIGGINS, and J. N. STRATHERN, 1990  Analysis of interchromosomal mitotic recombination. Curr. Genet. 18:29-39[Medline].

MCGILL, C. B., B. K. SHAFER, L. K. DERR, and J. N. STRATHERN, 1993  Recombination initiated by double-strand-breaks. Curr. Genet. 23:305-314[Medline].

MESELSON, M. S. and C. M. RADDING, 1975  A general model for genetic recombination. Proc. Natl. Acad. Sci. USA 72:358-361[Abstract/Free Full Text].

MODRICH, P., 1991  Mechanisms and biological effects of mismatch repair. Annu. Rev. Genet. 25:229-253[Medline].

MODRICH, P. and R. LAHUE, 1996  Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu. Rev. Biochem. 65:101-133[Medline].

NASMYTH, K. A., 1982  Molecular genetics of yeast mating-type. Annu. Rev. Genet. 16:439-500[Medline].

NELSON, J. R., C. W. LAWRENCE, and D. C. HINKLE, 1996  Thymine-thymine dimer bypass by yeast DNA polymerase . Science 272:1646-1649[Abstract].

NEGRITTO, M. T., X. WU, T. KUO, S. CHU, and A. M. BAILIS, 1997  Influence of DNA sequence identity on efficiency of targeted gene replacement. Mol. Cell. Biol. 17:278-286[Abstract].

NEUBERGER, M. S. and C. MILSTEIN, 1995  Somatic hypermutation. Curr. Opin. Immunol. 7:248-254[Medline].

PROLLA, T. A., D. M. CHRISTIE, and R. M. LISKAY, 1994a  Dual requirement in yeast DNA mismatch repair for MAL1 and PMS1, two homologs of the bacterial mutL gene. Mol. Cell. Biol. 14:407-415[Abstract/Free Full Text].

PROLLA, T. A., Q. PANG, E. ALANI, R. KOLODNER, and R. M. LISKAY, 1994b  MLH1, PMS1, and MSH2 interactions during the initiation of DNA mismatch repair in yeast. Science 265:1091-1093[Abstract/Free Full Text].

RADDING, C. M., 1982  Homologous pairing and strand exchange in genetic recombination. Annu. Rev. Genet. 16:405-437[Medline].

REENAN, R. A. G. and R. D. KOLODNER, 1992  Characterization of insertion mutations in the Saccharomyces cerevisiae MSH1 and MSH2 genes: evidence for separate mitochondrial and nuclear functions. Genetics 132:975-985[Abstract].

RESNICK, M. A., 1976  The repair of double-strand breaks in DNA: a model involving recombination. J. Theor. Biol. 59:97-106[Medline].

ROCHE, H., R. D. GIETZ, and B. A. KUNZ, 1994  Specificities of the Saccharomyces cerevisiae rad6, rad18, and rad52 mutators exhibit different degrees of dependence on the REV3 gene product, a putative nonessential/DNA polymerase. Genetics 140:443-456[Abstract].

SANTOS, M. E. and J. W. DRAKE, 1994  Rates of spontaneous mutation in bacteriophage T4 are independent of host fidelity determinants. Genetics 138:553-564[Abstract].

SCHAAPER, R. M., 1993  Base selection, proofreading and mismatch repair during DNA replication in Escherichia coli.. J. Biol. Chem. 268:23762-23765[Abstract/Free Full Text].

SCHWACHA, A. and N. KLECKNER, 1995  Identification of double Holliday junctions as intermediates in meiotic recombination. Cell 83:783-791[Medline].

SHERMAN, F., G. R. FINK and J. B. HICKS, 1986 Laboratory Course Manual for Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

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].

SZOSTAK, J. W., T. L. ORR-WEAVER, R. J. ROTHSTEIN, and F. W. STAHL, 1983  The double-strand- break repair model for recombination. Cell 33:25-35[Medline].

THALER, D. S., M. M. STAHL, and F. W. STAHL, 1987  Tests of the double-strand-break repair model for red-mediated recombination of phage lambda and plasmid lambda dv. Genetics 116:501-511[Abstract/Free Full Text].

THOMAS, D. C. and T. A. KUNKEL, 1991  Heteroduplex repair in extracts of human HeLa cells. J. Biol. Chem. 266:3744-3751[Abstract/Free Full Text].

TRAN, H. T., D. A. GORDENIN, and M. A. RESNICK, 1996  The prevention of repeat-associated deletions in Saccharomyces cerevisiae by mismatch repair depends on size and origin of deletions. Genetics 143:1579-1587[Abstract].

This article has been cited by other articles:

				M. O. Nowicki, R. Falinski, M. Koptyra, A. Slupianek, T. Stoklosa, E. Gloc, M. Nieborowska-Skorska, J. Blasiak, and T. Skorski BCR/ABL oncogenic kinase promotes unfaithful repair of the reactive oxygen species-dependent DNA double-strand breaks Blood, December 1, 2004; 104(12): 3746 - 3753. [Abstract] [Full Text] [PDF]

				A. J. Rattray, B. K. Shafer, C. B. McGill, and J. N. Strathern The Roles of REV3 and RAD57 in Double-Strand-Break-Repair-Induced Mutagenesis of Saccharomyces cerevisiae Genetics, November 1, 2002; 162(3): 1063 - 1077. [Abstract] [Full Text] [PDF]

				E. C. Friedberg, R. Wagner, and M. Radman Specialized DNA Polymerases, Cellular Survival, and the Genesis of Mutations Science, May 31, 2002; 296(5573): 1627 - 1630. [Abstract] [Full Text] [PDF]

				H. J. Bull, M.-J. Lombardo, and S. M. Rosenberg Stationary-phase mutation in the bacterial chromosome: Recombination protein and DNA polymerase IV dependence PNAS, July 17, 2001; 98(15): 8334 - 8341. [Abstract] [Full Text] [PDF]

				A. J. Rattray, C. B. McGill, B. K. Shafer, and J. N. Strathern Fidelity of Mitotic Double-Strand-Break Repair in Saccharomyces cerevisiae: A Role for SAE2/COM1 Genetics, May 1, 2001; 158(1): 109 - 122. [Abstract] [Full Text]

				F. Paques and J. E. Haber Multiple Pathways of Recombination Induced by Double-Strand Breaks in Saccharomyces cerevisiae Microbiol. Mol. Biol. Rev., June 1, 1999; 63(2): 349 - 404. [Abstract] [Full Text] [PDF]