EXO1 and MSH6 Are High-Copy Suppressors of Conditional Mutations in the MSH2 Mismatch Repair Gene of Saccharomyces cerevisiae

Corresponding author: Eric Alani, Department of Molecular Biology and Genetics, Cornell University, 459 Biotechnology Bldg., Ithaca, NY 14853-2703., eea3{at}cornell.edu (E-mail)

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In Saccharomyces cerevisiae, Msh2p, a central component in mismatch repair, forms a heterodimer with Msh3p to repair small insertion/deletion mismatches and with Msh6p to repair base pair mismatches and single-nucleotide insertion/deletion mismatches. In haploids, a msh2 mutation is synthetically lethal with pol3-01, a mutation in the Pol proofreading exonuclease. Six conditional alleles of msh2 were identified as those that conferred viability in pol3-01 strains at 26° but not at 35°. DNA sequencing revealed that mutations in several of the msh2^ts alleles are located in regions with previously unidentified functions. The conditional inviability of two mutants, msh2-L560S pol3-01 and msh2-L910P pol3-01, was suppressed by overexpression of EXO1 and MSH6, respectively. Partial suppression was also observed for the temperature-sensitive mutator phenotype exhibited by msh2-L560S and msh2-L910P strains in the lys2-Bgl reversion assay. High-copy plasmids bearing mutations in the conserved EXO1 nuclease domain were unable to suppress msh2-L560S pol3-01 conditional lethality. These results, in combination with a genetic analysis of msh6 pol3-01 and msh3 pol3-01 strains, suggest that the activity of the Msh2p-Msh6p heterodimer is important for viability in the presence of the pol3-01 mutation and that Exo1p plays a catalytic role in Msh2p-mediated mismatch repair.

DNA mispairs can result from polymerase errors that occur during replication. These errors include nucleotide misincorporations and insertion/deletion mutations that are thought to arise as a result of misalignment of template and primer strands within tracts of short repeat sequences (STREISINGER et al. 1966 ; STRAND et al. 1993 ). The replication errors must be removed prior to the next round of replication to prevent fixation of the mutation in the genome. In the yeast Saccharomyces cerevisiae, the Pol (encoded by POL2) and Pol (encoded by POL3) processive DNA replication polymerases each contain an intrinsic proofreading function. This activity is capable of 3' to 5' exonucleolytic cleavage of nascent strands containing replication errors (KESTI and SYVAOJA 1991 ; MORRISON et al. 1991 ; SIMON et al. 1991 ). Other repair pathways, including the MSH2-dependent mismatch repair (MMR) pathway, can also recognize and repair replication errors, thus increasing replication fidelity (SCHAAPER 1993 ; KOLODNER and MARSISCHKY 1999 ).

The best understood MMR system is the Escherichia coli mutHLS repair pathway (reviewed in MODRICH and LAHUE 1996 ). MMR is thought to be initiated by MutS binding to a mispair. Interactions between MutL, the MutS-mispair complex, and ATP are thought to result in the formation of a mismatch-MutS-MutL complex. This complex activates the methylation-sensitive endonuclease MutH, which cleaves the newly synthesized strand at hemimethylated sites either 5' or 3' of the mismatch. In yeast, six MutS (MSH1-6) and four MutL (PMS1 and MLH1-3) homologs have been identified (reviewed in MODRICH and LAHUE 1996 ; MODRICH 1997 ; KOLODNER and MARSISCHKY 1999 ). Msh2p-Msh3p and Msh2p-Msh6p heterodimers form the initiation complexes for two partially redundant repair pathways that act to repair replication errors. The Msh2p-Msh3p heterodimer primarily corrects small loop mismatches while the Msh2p-Msh6p heterodimer primarily corrects nucleotide substitutions and small loop mismatches. Binding of mismatches by these heterodimers is thought to signal recruitment of a heterodimer of MutL homologs. The Mlh1p-Pms1p heterodimer functions in both the Msh3p- and Msh6p-dependent repair pathways. Another heterodimer, Mlh1p-Mlh3p, has also been implicated in the Msh3p-dependent repair pathway (FLORES-ROZAS and KOLODNER 1998 ; WANG et al. 1999 ). No MutH homolog has been identified in eukaryotes; however, the replication processivity factor proliferating cell nuclear antigen (PCNA; encoded by POL30) has been implicated in MMR at steps prior to and during strand resynthesis and may play a critical role in strand discrimination by targeting MMR proteins to excise newly synthesized DNA (JOHNSON et al. 1996 ; UMAR et al. 1996 ; GU et al. 1998 ; CHEN et al. 1999 ).

In MMR, following mismatch recognition and strand discrimination, the nascent strand is excised. In E. coli excision is accomplished by Helicase II and the single-stranded exonuclease Exo1 (3'-5' excision), ExoVII, or Rec J (5'-3' excision; MODRICH and LAHUE 1996 ). In yeast, Exo1p, a nuclease belonging to the Rad27p/FEN-1 family of double-stranded DNA 5' to 3' exonucleases, has been implicated in MMR. Exo1p was first linked to MMR through a two-hybrid interaction with Msh2p (TISHKOFF et al. 1997A ). An exo1 strain exhibits mutation rates that are much lower than in other MMR mutants, suggesting that other exonucleases with redundant functions can also act in the repair process or that Exo1p plays a minor role in MMR (TISHKOFF et al. 1997A ). S. cerevisiae and Schizosaccharomyces pombe exo1 mutants also display defects in mitotic and meiotic recombination (SZANKASI and SMITH 1995 ; FIORENTINI et al. 1997 ; TISHKOFF et al. 1997A ). In exo1 strains, the mitotic recombination rate between nontandem direct repeat sequences is reduced sixfold (FIORENTINI et al. 1997 ). Finally, Exo1p appears to act in the repair of UV-induced damage via a repair pathway that is independent from nucleotide excision and MMR (QIU et al. 1998 ).

Current models predict that replication errors are first acted upon by DNA polymerase activities but those that escape proofreading then become substrates for MMR, i.e., that the exonuclease functions of DNA polymerases and MMR act in series (MORRISON et al. 1993 ). Consistent with this model, pol3-01, a mutation in the proofreading exonuclease of Pol, is synthetically lethal with the MMR or DNA replication mutations msh2, pms1, exo1, rfc1, rad27, pol30-52, and pol2-04 in haploid strains (MORRISON et al. 1993 ; MORRISON and SUGINO 1994 ; KOKOSKA et al. 1998 ; TRAN et al. 1999B ; XIE et al. 1999 ). pol3-01 mutants display an increased rate of nucleotide misincorporations as well as instability of short repeat tracts (MORRISON et al. 1993 ; MORRISON and SUGINO 1994 ; TRAN et al. 1999B ). With the exception of the rad27 mutation, diploid strains homozygous for the pol3-01 and DNA replication or MMR mutations are viable and display synergistic mutation rates, consistent with these pathways acting in series.

This study focuses on the genetic interactions of Msh2p with repair and replication factors. We took advantage of the msh2 pol3-01 synthetic lethality to identify six temperature-sensitive msh2 alleles. Overexpression of EXO1 and MSH6 suppressed the inviability of msh2-L560S pol3-01 and msh2-L910P pol3-01 strains, respectively. The results suggest that the interaction between defects in MMR and the exonuclease activity of Pol is specific to the Msh2p-Msh6p repair pathway and that the nuclease activity of Exo1p plays a catalytic role in the Msh2p-Msh6p MMR process.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Media, reagents and chemicals:
Yeast strains were grown in either YPD or minimal selective media (ROSE et al. 1990 ). Selective media contained 0.7% yeast nitrogen base, 2% agar, and 0.09% of a drop-out mix that lacks the appropriate amino acid; 2% glucose, 2% sucrose, and 2% galactose were included as indicated. When required, canavanine (Sigma, St. Louis) was included in minimal selective media lacking arginine at 60 mg/liter. Plates of 5-fluoroorotic acid (5-FOA; US Biological, San Antonio, TX) were prepared as described (ROSE et al. 1990 ).

Genetic procedures:
Yeast were transformed with episomal vectors using the lithium acetate method described by GIETZ and SCHIESTL 1991 . Tetrad dissections (ROSE et al. 1990 ) to test for synthetic lethality between pol3-01 and msh6 or msh3 mutations were performed using diploids derived from crosses of EAY575 (MAT pol3-01 ura3-52 leu21 trp163) with EAY337 (MATa msh6::hisG ura3-52 leu21 trp163) or with EAY420 (MATa msh3::hisG ura3-52 leu21 trp163). All yeast strains used in the tetrad analysis and mutation rate studies (Table 1 and Table 2) were derived from the S288C background (WINSTON et al. 1995 ). Allele tests were performed using PCR-based assays as described previously (XIE et al. 1999 ).

Nucleic acid techniques:
Oligonucleotide synthesis and double-stranded DNA sequencing was performed at the Cornell Biotechnology Analytical-Synthesis Facility (Ithaca, NY). All restriction endonucleases, T4 DNA ligase, and Vent polymerase were from New England Biolabs (Beverly, MA) and used according to the manufacturer's specifications. Plasmid DNA was isolated using a QiaPrep Spin kit from QIAGEN (Valencia, CA) and all DNA manipulations were performed as described previously (MANIATIS et al. 1982 ).

MSH2 and EXO1 mutagenesis:
The S. cerevisiae strain RKY2151 (MATa ade2 leu2-3,112 his31 msh2::hisG pol3-01 trp1-289 ura3-52 + pMSH2 URA3 ADE2 ARSH4 CEN6) was used to identify conditionally lethal msh2 alleles and was kindly provided by R. Kolodner. The library used to identify conditional msh2 alleles was created using PCR-generated mutagenesis of the entire MSH2 open reading frame as described (STUDAMIRE et al. 1999 ). Briefly, 2 µg of pEAA38 (MSH2, URA3, ARSH4, CEN6; ALANI et al. 1997 ) was used as template in 12 separate 100-µl PCR reactions using standard concentrations of Taq polymerase, buffers, and primers as recommended by the manufacturer (Perkin Elmer-Cetus, Norwalk, CT). Each PCR reaction involved 12 cycles using a 1-min denaturation step at 94°, a 1-min annealing step at 50° or 55°, and a 2-min polymerization step at 72°. Reactions used either primers AO143 (5' GCAAGTGTAGCGGTCACGC) and AO39 (5' GTTAATTTCAGTTAGCGG) to amplify a 3.3-kb 5' fragment of MSH2 or primers AO144 (5' AGTCAGTGAGCGAGGAAGC) and AO7 (5' GGAAACTTAGAGGATGTC) to amplify a 2-kb 3' fragment of MSH2. The amplified fragments were gel purified, digested with restriction enzymes, and subcloned into the corresponding sites of pEAA54 (MSH2, TRP1, ARSH4, CEN6; STUDAMIRE et al. 1999 ). The 5' MSH2 fragment was cut with NotI and BglII to give an ~2.8-kb insert and the 3' MSH2 fragment was cut with KpnI and BglII to yield an ~1.3-kb insert. A total of 1 µl of each ligation was used to transform E. coli by electroporation and each transformation yielded ~3000 transformants containing PCR-mutagenized pEAA54.

The exo1-D171A and exo1-D173A alleles were created using overlapping PCR site-directed mutagenesis (HO et al. 1989 ). DNA sequencing of the entire insert did not reveal any other mutations in these constructs. GAL1 activation domain fusions of exo1-D171A (pEAE140) and exo1-D173A (pEAE142) used in the co-immunoprecipitation studies were created by subcloning an AvrII-XhoI fragment containing the respective mutation into pRDK502 (GAL1-B42-EXO1, 2µ, TRP1; kindly provided by R. Kolodner; TISHKOFF et al. 1997A ).

Overexpression plasmids:
The following plasmids were used in the high-copy suppression studies described in Fig 2 and Fig 4 and Table 1. pRS425 (LEU2, 2µ; CHRISTIANSON et al. 1992 ) served as a control plasmid for all of the studies. pEAM67 (EXO1, LEU2, 2µ) and mutant derivatives pEAM69 (exo1-D171A, LEU2, 2µ) and pEAM71 (exo1-D173A, LEU2, 2µ) were all derived from pRDK480 (EXO1, LEU2, 2µ; TISHKOFF et al. 1997A ). pEAM66 (RAD27, LEU2, 2µ) and pEAE108 (GAL1-POL30, LEU2, 2µ) were derived from pR2.14 (RAD27, 2µ, URA3; kindly provided by L. Prakash; SOMMERS et al. 1995 ) and pCH1525 (GAL1-POL30, CEN6, ARSH4, LEU2; kindly provided by C. Holm; MCALEAR et al. 1996 ), respectively. pEAE107 (GAL10-MLH1, GAL1-PMS1, LEU2, 2µ) was derived from pBL19 (GAL1-PMS1, 2µ, URA3; kindly provided by R. Lahue) and pEAE49 (GAL10-MLH1, 2µ, TRP1; Alani laboratory collection). pEAE127 (GAL10-MSH2, LEU2, 2µ), pEAE82 (GAL10-MSH3, LEU2, 2µ), and pEAE80 (GAL10-MSH6, LEU2, 2µ) were constructed in the Alani laboratory and are similar to plasmids described in BOWERS et al. 1999 .

View larger version (47K): [in this window] [in a new window] [Download PPT slide]   Figure 1. msh2 alleles display temperature-sensitive synthetic lethality with pol3-01. (A) Amino acid substitutions in the six temperature-sensitive msh2 alleles identified from a msh2 PCR-mutagenized library for conditional synthetic lethality with pol3-01. (B) Growth phenotypes for msh2ts alleles expressed from ARS CEN plasmids in RKY2151 (msh2, pol3-01 + pMSH2,URA3) at permissive (26°) and restrictive (35°) temperatures on 5-FOA-containing minimal media. (C) Immunoblots of overexpressed msh2ts alleles. EAY281 (msh2) was transformed with pEAE132 (lane 1, msh2-L910P), pEAE133 (lane 2, msh2-L560S), pEAE134 (lane 3, msh2-V45D), pEAE136 (lane 4, msh2-L531S, L642S), pEAE137 (lane 5, msh2-L393S, K489G), pEAE138 (lane 6, msh2-V57D, F232S, K498E), pEAE20 (lane 7, MSH2), and vector only (lane 8). Transformants were grown and prepared for Western blot analysis using the Msh2p polyclonal antibody as described in MATERIALS AND METHODS. Purified Msh2p was included as marker (lane 9).

View larger version (48K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Suppression of synthetic lethality of msh2ts alleles with pol3-01. (A) RKY2151 (msh2, pol3-01 + pMSH2, URA3) bearing the indicated pEAA54-derived msh2ts plasmids was transformed with pEAE127 (GAL10-MSH2, LEU2, 2µ), pEAE82 (GAL10-MSH3, LEU2, 2µ), pEAE80 (GAL10-MSH6, LEU2, 2µ), pEAE107 (GAL10- MLH1, GAL1-PMS1, LEU2, 2µ), pEAE108 (GAL1-POL30, LEU2, 2µ), or pRDK480 (EXO1, LEU2, 2µ). Strains were plated onto minimal yeast plates containing 5-FOA and 2% galactose, 2% sucrose at 26° and 35°. Y and N indicate viable and inviable at 35°, respectively. (B) RKY2151 containing msh2-L560S (top) or msh2-L910P (bottom) was transformed with the indicated overexpression plasmids and plated onto 2% galactose, 2% sucrose, and 5-FOA-containing minimal plates at 26° and 35°.

View larger version (79K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Msh2p-Exo1p co-immunoprecipitation. (A) 12CA5 anti-HA antibody was used to immunoprecipitate HA-tagged Exo1p from cell extracts as described in MATERIALS AND METHODS. The presence of Msh2p in immunoprecipitates was detected in immunoblots using polyclonal Msh2p antibody. Crude extracts and immunoprecipitations were prepared from EGY48 overexpressing LexA-Msh2p and Exo1p (lane 1), LexA-Msh2p and exo1-D171Ap (lane 2), LexA-Msh2p and exo1-D173Ap (lane 3), LexA-Msh2p only (lane 4), LexA-Msh2p and Exo1p (lane 6), and LexA-msh2-L560Sp and Exo1p (lane 7). Strains were grown at 26° for lanes 1–3 and 35° for lanes 4–7. The same crude extract used in lane 6, prepared from EGY48 expressing LexA-Msh2p and Exo1p, was used in a reaction containing protein-G-Agarose but no 12CA5 antibody to assess nonspecific interactions of the agarose beads (lane 5). (B) Membrane in A was stripped and reprobed using 12CA5 antibody as control for Exo1p expression and immunoprecipitation.

View larger version (66K): [in this window] [in a new window] [Download PPT slide]   Figure 4. Alanine substitutions in the Exo1p nuclease domain disrupt suppression of msh2-L560S pol3-01. (A) Amino acid alignment of a portion of the conserved nuclease domain of Rad27p/FEN-1 homologs (GARY et al. 1999 ). Highly conserved residues are shown with diamonds. Arrows indicate the location of the exo1-D171A and exo1-D173A alleles. (B) RKY2151 (msh2, pol3-01 + pMSH2,URA3) containing pEAA101 (msh2-L560S) was transformed with pEAM67 (EXO1, LEU2, 2µ), pEAM71 (exo1-D173A, LEU2, 2µ), pEAM69 (exo1-D171A, LEU2, 2µ), pEAM66 (RAD27, LEU2, 2µ), pEAE127 (GAL10-MSH2, LEU2, 2µ), or pRS425 (LEU2, 2µ). Transformants were plated onto 2% galactose, 2% sucrose, and 5-FOA-containing minimal media at 26° and 35°.

Determination of mutation rates:
Reversion of lys2-Bgl to Lys⁺ was examined in both EAY603 (MAT leu21 msh2::hisG lys2-BglII trp163 ura3-52) and EAY620 (MAT leu21 lys2-BglII trp163 ura3-52 his3200 exo1::HIS3 msh2::hisG). The forward mutation rate to canavanine resistance (REENAN and KOLODNER 1992 ) was measured in EAY541 or EAY542 (MATa ade2 cyh^S his3 msh2::hisG trp1 ura3). All steps were performed at the indicated temperature (26° or 35°). These strains were transformed with the following vectors: pRS414 (TRP1, ARSH4, CEN6; SIKORSKI and HIETER 1989 ), pEAA54 (MSH2, TRP1, ARSH4, CEN6), and pEAA54 derivatives pEAA97 (msh2-V45D), pEAA98 (msh2-L393S, K489G), pEAA99 (msh2-V57D, F232S, K498E), pEAA100 (msh2-L910P), pEAA101 (msh2-L560S), and pEAA104 (msh2-L531S, L642S). Transformants were streaked to single colonies on selective minimal media. For lys2-Bgl reversion, single colonies were used to set up 11 independent minimal media cultures. A total of 5.5 ml of each overnight culture was plated to lysine drop-out media and dilutions of the same culture were plated to permissive media. Canavanine assays were performed as described (REENAN and KOLODNER 1992 ; BOWERS et al. 1999 ). The median rate of lys2-Bgl reversion or canavanine resistance was determined using the method of LEA and COULSON 1949 .

Protein stability studies:
Immunoblots were performed using crude extracts of EAY281 (MATa ura3-52 leu21 trp163 msh2::hisG) transformed with pEAE20 (GAL10-MSH2, URA3, 2µ; ALANI et al. 1995 ) derived plasmids containing the conditional msh2 alleles. Transformants were grown in 5-ml overnight cultures containing 2% galactose and 2% sucrose at the permissive or restrictive temperatures of 22° and 35°, respectively. Crude extracts were prepared by resuspending cell pellets in 250 µl 150 mM NaCl buffer A (25 mM Tris, pH 7.5, 1 mM EDTA, 10 mM ß-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride). A total of 250 µl of glass beads was added to each tube and the cells were vortexed in an Eppendorf (Westbury, NY) model 5432 platform vortexer at 4° for 30 min. Total protein was determined by Bradford assay (BRADFORD 1976 ; Bio-Rad Protein Assay, Hercules, CA) and ~35 µg total protein was loaded for each sample on 8% SDS-PAGE. Gels were transferred to nitrocellulose using a Bio-Rad Trans-Blot SD Semi-Dry Electrophoretic Transfer Cell. Immunoblotting was performed using a rabbit Msh2p polyclonal antibody at a 1:2000 dilution and ECL chemiluminescence (Amersham, Arlington Heights, IL).

Msh2p-Exo1p co-immunoprecipitation studies:
EGY48 (MAT his3 leu2::3Lexop-LEU2 ura3 trp1 LYS2) was transformed with combinations of the following plasmids: pRDK502 (GAL1-HA-EXO1, TRP1, 2µ; TISHKOFF et al. 1997A ), pRDK584 (LexA-Hairy, HIS3, 2µ; kindly provided by R. Kolodner), pRDK371 (LexA-MSH2, HIS3, 2µ; TISHKOFF et al. 1997A ), pEAM59 (LexA-msh2-L560S, HIS3, 2µ), pEAE140 (GAL1-HA-exo1-D171A, TRP1, 2µ), and pEAE142 (GAL1-HA-exo1-D173A, TRP1, 2µ). Cultures of 50 ml were grown at the indicated temperature in drop-out media containing 2% galactose and 2% sucrose to an OD₆₀₀ of 1.5–2. Crude extracts were prepared as described in 150 mM NaCl buffer A (ALANI 1996 ). Following lysis, Triton X-100 was added to a 1% final concentration. All immunoprecipitations were performed at 4°. The reactions were precleared by incubation of 500 µl of crude extract with 20 µl protein-G-Agarose diluted 1:1 in 150 mM NaCl buffer A, 1% Triton X-100 for 30 min. Samples were then centrifuged at 3000 rpm. The crude extract was transferred to new tubes and incubated with 10 µg of 12CA5 anti-HA antibody for 1 hr. A total of 20 µl protein-G-Agarose was then added and the reaction was incubated for another hour. Reactions were centrifuged at 3000 rpm and the crude extract was removed. The agarose beads were then washed four times with 500 µl 150 mM NaCl buffer A, 1% Triton X-100. Immunoblots were performed as described above using polyclonal Msh2p antibody and ECL, followed by stripping and reprobing using monoclonal 12CA5 anti-HA antibody.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of msh2 alleles that exhibit conditional lethality with pol3-01:
As outlined in the Introduction, haploid yeast strains that contain both the pol3-01 and msh2 mutations are inviable (TRAN et al. 1999B ). Using a plasmid shuffle approach, we identified msh2 alleles that conferred temperature-sensitive inviability in the pol3-01 msh2 haploid strain RKY2151 that contains MSH2 on a URA3 single copy vector. RKY2151 was transformed with a PCR-mutagenized MSH2 TRP1 library and transformants were selected for loss of the MSH2 URA3 plasmid by replica plating to 5-FOA plates (ROSE et al. 1990 ; MATERIALS AND METHODS). Viability of the transformants on 5-FOA plates was assessed at 16°, 26°, and 35°. Approximately 19,000 transformants from 12 individually mutagenized pools were screened. After recovery of the initial candidates and retransformation into RKY2151 to confirm the growth phenotype, six temperature-sensitive alleles (msh2^ts) were identified from 6 of the mutagenized pools. As shown in Fig 1B, when transformed with an MSH2 TRP1 control plasmid, RKY2151 is viable on media containing 5-FOA at 26° and 35°. However, this strain does not grow at either temperature when transformed with the TRP1 vector. In combination with pol3-01, the six msh2^ts alleles exhibit only minor growth defects at 26° while displaying inviability at 35°. No cold-sensitive alleles were obtained.

Msh2p shares strong conservation with other MutS homologs (CROUSE 1998 ; KOLODNER and MARSISCHKY 1999 ). Two highly conserved regions of Msh2p with identified functions include an ATP binding domain and an Msh6p dimerization domain (ALANI 1996 ; ALANI et al. 1997 ; GUERRETTE et al. 1998 ). Sequencing revealed that each of the mutations in the msh2^ts alleles is located in a conserved amino acid in the MSH homologs (Fig 1A; data not shown). Mutations in five of the alleles mapped to amino acids outside the ATP binding and Msh6p dimerization domains. The msh2-L910P mutation is located within the Msh6p dimerization domain (ALANI 1996 ; ALANI et al. 1997 ). The mutations in msh2-L393S, K489G were separated by subcloning. Both single mutant alleles were viable in combination with pol3-01 at the restrictive temperature (data not shown), suggesting that lethality requires the presence of both msh2 mutations. The msh2-L560S allele was isolated twice from the same PCR pool while msh2-L910P was isolated from two independent PCR pools.

msh2^ts alleles exhibit temperature-sensitive lys2-Bgl reversion:
To determine if msh2 alleles temperature sensitive for synthetic lethality were also temperature sensitive for MMR, the six msh2 alleles were tested for mutator phenotypes in the absence of the pol3-01 mutation at permissive and restrictive temperatures in the lys2-Bgl reversion and canavanine resistance assays. The lys2-Bgl reversion assay makes use of a strain with a four-nucleotide insertion that creates a frameshift mutation in the LYS2 gene. If a compensatory frameshift mutation occurs the strain will be able to grow on media lacking lysine. Although reversion to Lys⁺ can occur within an ~150-bp region, sequencing has revealed that in MMR-defective strains reversion results primarily from one-nucleotide deletions at one of three short mononucleotide repeat sequences (MARSISCHKY et al. 1996 ; GREENE and JINKS-ROBERTSON 1997 ; TISHKOFF et al. 1997B ; FLORES-ROZAS and KOLODNER 1998 ). An msh2 strain exhibited an average reversion rate 65-fold higher than wild type at 26° and 90-fold higher than wild type at 35° (Table 1). The single mutants msh2-L560S and msh2-L910P exhibited a strong temperature-sensitive phenotype, with a reversion rate at 35° that was 15- to 20-fold higher than the rate at 26°. The double mutants msh2-L393S, K489G and msh2-L531S, L642S displayed a weaker temperature-sensitive phenotype with an ~5.5-fold higher reversion rate at 35° compared to 26°. The msh2-V45D and msh2-V57D, F232S, K498E alleles did not exhibit temperature sensitivity in this assay.

In contrast, none of the msh2^ts alleles displayed an intrinsic temperature sensitivity for the rate of forward mutations in the CAN1 gene. These mutations confer resistance to the toxic arginine analog canavanine (Table 2). In msh2 strains, forward mutations in CAN1 include nucleotide misincorporations and single-nucleotide deletions (MARSISCHKY et al. 1996 ). In the canavanine resistance assay, an msh2 strain exhibited a mutation rate 15- to 20-fold higher than wild type. Five of the msh2^ts alleles displayed an average mutation rate similar to a null phenotype at both temperatures. However, the msh2-L560S and msh2-V57D, F232S, K498E alleles displayed more moderate phenotypes at both temperatures.

Stability of msh2p proteins:
One explanation for temperature sensitivity is that a mutation destabilizes the protein, resulting in degradation at the restrictive temperature. Immunoblotting was performed to assess steady-state protein levels of the msh2^ts alleles at the permissive temperature of 22° and the nonpermissive temperature of 35°. Because Msh2p expressed from a single copy ARS CEN plasmid is at the threshold of detection with our Msh2p polyclonal antibody, the six msh2^ts alleles were expressed in high copy from the GAL10 promoter in an msh2 strain (Fig 1C; see MATERIALS AND METHODS). Five of the alleles displayed protein levels similar to wild type at both permissive and nonpermissive temperatures, suggesting that temperature sensitivity is not due to protein instability at 35°. Full-length protein was not detected for the msh2-V45D allele at either temperature. However, a polypeptide of ~40 kD was observed at both temperatures (data not shown). This raises the possibility that the msh2-V45D mutation destabilizes Msh2p resulting in a specific degradation product and the presence of this degradation product confers temperature sensitivity. Alternatively, full-length protein might be present at the permissive temperature but at levels lower than can be detected in this assay.

msh2-L910P disrupts Msh2p-Msh6p interactions:
To further understand the defects in the msh2^ts alleles, we examined suppression of msh2^ts pol3-01 synthetic lethality by overexpression of MMR factors. These studies were conducted by transforming the msh2 pol3-01 strain carrying the MSH2 URA3 plasmid with each temperature-sensitive msh2 allele and overexpression plasmids carrying the following MMR genes: GAL10-MSH2, GAL10-MSH3, GAL10-MSH6, GAL10-MLH1-GAL1-PMS1, GAL1-POL30 (PCNA), and EXO1 (MATERIALS AND METHODS). These genes were chosen as candidates for dosage suppression as each has been implicated in Msh2p-dependent MMR and they are thought to interact directly or indirectly with Msh2p itself (KOLODNER and MARSISCHKY 1999 ). Transformants containing the overexpression plasmids were plated to media containing 5-FOA to select for loss of the MSH2 URA3 plasmid at permissive and restrictive temperatures. Consistent with the recessive nature of the msh2^ts alleles, all were suppressed by overexpression of MSH2 (Fig 2A). Two cases of allele-specific suppression were identified (Fig 2A and see below).

The msh2-L910P mutation is located within an Msh2p-Msh6p interaction domain. We have demonstrated previously that mutations in this domain of Msh2p confer Msh6p interaction defects as well as defects in in vivo MMR assays (ALANI 1996 ; ALANI et al. 1997 ). Consistent with these previous findings, msh2-L910P exhibited a defect in Msh2p-Msh6p subunit interactions that was detected during the attempted purification of an msh2-L910Pp-Msh6p complex (data not shown; ALANI 1996 ). The msh2-L910P allele was suppressed in an allele-specific manner by overexpression of MSH6 (Fig 2B), suggesting that the defect in heterodimer formation is linked to synthetic lethality of this allele. In addition, overexpression of MSH6 reduced the rate of lys2-Bgl reversion in msh2-L910P strains twofold (Table 1). Together, these findings suggested that msh6 mutations would also display synthetic lethality in a pol3-01 strain. To test this we crossed an msh6 strain (EAY337) with a pol3-01 strain (EAY575) and sporulated the heterozygous diploid. Spores derived from dissecting msh6/MSH6 pol3-01/POL3 tetrads displayed reduced viability, with 24 out of 35 tetrads displaying only 3 viable spores. The overall spore viability in this cross was 71%, which is very close to the 25% reduction in spore viability expected for synthetic lethality of unlinked genes. Allele tests of 12 viable spores from tetrads revealed that none carried both mutant alleles.

Because Msh2p also forms a heterodimer with Msh3p, we examined msh3 pol3-01 mutations for viability. We crossed an msh3 strain (EAY420) with a pol3-01 strain (EAY575) and dissected tetrads. In this cross 30 out of 35 tetrads had four viable spores. Overall spore viability was maintained at 93%. In addition, allele tests identified viable msh3 pol3-01 spores. These data indicate to us that pol3-01 is not synthetically lethal with an msh3 and that synthetic lethality between MMR mutations and pol3-01 is limited to defects in the Msh2p-Msh6p repair pathway.

msh2-L560S is suppressed by EXO1 overexpression:
As shown in Fig 2B, the synthetic lethality of the msh2-L560S allele was suppressed in an allele-specific manner by overexpression of EXO1. Additionally, EXO1 overexpression partially suppressed the mutator phenotype exhibited by this allele at 35° in the lys2-Bgl reversion assay (Table 1). Exo1p was first linked to MMR through a two-hybrid interaction with Msh2p and the interaction of these proteins was confirmed by co-immunoprecipitation of Exo1p and Msh2p from the two-hybrid strains (TISHKOFF et al. 1997A ). We made use of these same assays to determine if msh2-L560S exhibits a defect for Exo1p interaction. As shown in Fig 3, in the two-hybrid strain grown at 35° this allele is expressed and co-immunoprecipitates with Exo1p to a level similar to wild type. A deletion of amino acids 518–643 in Msh2p (msh2-518-643; ALANI 1996 ) also did not disrupt the Msh2p-Exo1p interaction, indicating that the msh2-L560S allele is probably not located in an Exo1p interaction domain (data not shown). In addition, we observed that an msh2-L560Sp-Msh6p complex purified like the wild-type complex, indicating that msh2-L560Sp was not defective in Msh2p-Msh6p subunit interactions (data not shown).

One possible explanation for suppression of msh2-L560S pol3-01 synthetic lethality by overexpression of EXO1 is that Exo1p plays a structural role in stabilizing interactions of msh2-L560Sp with other components of MMR or replication. Alternatively, the catalytic activity of the nuclease is required. To address this issue, alanine substitutions were created in aspartic acid residues of the Exo1p nuclease domain conserved between Exo1p and Rad27p/FEN-1 family members (Fig 4A). Alanine substitutions at Asp-181 in human FEN-1 and Asp-179 in yeast Rad27p (rad27-n) have been shown previously to disrupt nuclease activity without disrupting DNA substrate binding (SHEN et al. 1996 ; GARY et al. 1999 ). Overexpression of the exo1-D171A and exo1-D173A alleles from the native promoter on 2µ plasmids did not result in suppression of msh2-L560S pol3-01 synthetic lethality (Fig 4B), suggesting that suppression requires a functional Exo1p nuclease. While overexpression of the rad27-n allele in wild-type yeast resulted in a dominant negative growth defect (GARY et al. 1999 ), in the canavanine resistance assay the exo1-D171A and exo1-D173A alleles did not complement an exo1 and did not exhibit a dominant negative phenotype when overexpressed in a wild-type strain (data not shown). To assess the exo1 alleles' ability to interact with Msh2p, the alleles were cloned into pRDK502, which expresses a B42 activation domain-HA epitope-EXO1 fusion under control of the GAL1 promoter (TISHKOFF et al. 1997A ). Both the exo1-D171Ap and the exo1-D173Ap proteins displayed wild-type levels of overexpression and interaction with Msh2p as measured by co-immunoprecipitation (Fig 3, lanes 1–3; data not shown).

The suppression of msh2-L560S pol3-01 synthetic lethality by overexpression of EXO1 could be due to a catalytic function but might not directly relate to the role of Exo1p in MMR. Exo1p functions in several repair pathways and is believed to have overlapping functions with Rad27p, a homologous protein involved in Okazaki fragment processing during lagging strand synthesis through removal of the final 5' ribonucleotide of the RNA primer following the activity of RNase HI (BAMBARA et al. 1997 ; LIEBER 1997 ; KOLODNER and MARSISCHKY 1999 ). While Rad27p is not thought to function in Msh2p-dependent MMR, pol3-01 and rad27 are also synthetically lethal (KOKOSKA et al. 1998 ). Additionally, overexpression of EXO1 suppresses rad27 alleles for growth and DNA repair defects (TISHKOFF et al. 1997A ; Y. XIE and E. ALANI, unpublished data). If EXO1 overexpression is suppressing a defect other than Msh2p-dependent MMR, then overexpression of RAD27 might also be capable of suppressing the synthetic lethality of msh2-L560S pol3-01. As shown in Fig 4B, overexpression of RAD27 did not suppress msh2-L560S pol3-01 synthetic lethality, suggesting that the Rad27p and Exo1p exonucleases are not redundant for this activity. This is consistent with suppression reflecting a specific MMR activity by the Exo1p nuclease.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this study we identified six msh2^ts alleles that were conditional for synthetic lethality with the pol3-01 mutation. The majority of these mutations map to regions of Msh2p with unidentified functions. Overexpression of the MSH6 and EXO1 genes resulted in allele-specific suppression of the msh2-L910P and msh2-L560S mutations, respectively. These data in combination with other work presented suggest that defects in the Msh2p-Msh6p-dependent repair pathway result in synthetic lethality with pol3-01 and that Exo1p plays a catalytic role in MMR.

EXO1 overexpression suppresses synthetic lethality via a catalytic function:
In the lys2-Bgl reversion assay an exo1 strain exhibited a lys2-Bgl reversion rate that was only 3-fold higher than wild type; in the same assay an msh2 strain displayed a 65- to 90-fold higher reversion rate (Table 1). An exo1 msh2 double mutant displayed a reversion rate similar to that observed for the msh2 strain, consistent with both proteins playing a role in MMR (Table 1). Similar conclusions have been reached in other mutator assays involving the exo1 mutation (TISHKOFF et al. 1997A ; TRAN et al. 1999A , TRAN et al. 1999B ).

Our observations are consistent with Exo1p playing an important role in MMR. First, we observed that EXO1 overexpression can specifically suppress both the conditional lethality observed in msh2-L560S pol3-01 mutants and the conditional mutator phenotype observed in msh2-L560S mutants (Fig 2). Second, the failure to observe suppression of msh2-L560S pol3-01 synthetic lethality by overexpressing the exo1 nuclease mutations suggests that Exo1p plays a specific catalytic role in the suppression of msh2-L560S pol3-01 synthetic lethality rather than a structural role (Fig 4). Third, suppression by EXO1 overexpression also does not appear to be due to functional redundancy of Exo1p with Rad27p. Functional redundancy for Exo1p and Rad27p in DNA replication has been suggested by suppression studies that showed that some of the rad27 replication defects can be overcome by overexpression of EXO1 (TISHKOFF et al. 1997A ; Y. XIE and E. ALANI, unpublished results). Furthermore, these two proteins share strong sequence conservation, especially within the nuclease domain (TISHKOFF et al. 1997A ; QIU et al. 1999 ). However, unlike MMR mutants, rad27 mutants display a mutation spectra of primarily larger deletions and duplications and msh2 rad27 strains exhibit a synergistic mutation rate that is inconsistent with a role for Rad27p in MMR (TISHKOFF et al. 1997B ). Our finding that msh2-L560S pol3-01 synthetic lethality was not suppressed by overexpression of RAD27 provides additional evidence that Exo1p suppression of the msh2-L560S mutation is MMR specific.

Finally, an analysis of previously identified msh2 mutations is consistent with a role for Exo1p in MMR. The msh2-L560S mutation is located in the same region of Msh2p as three msh2 alleles (msh2-G561D, -K564E, -G566D) that confer a dominant negative mutator phenotype when overexpressed in wild-type strains. These alleles cause strong defects in MMR but confer wild-type function in a double-strand break repair (DSBR) assay. In this assay the Msh2p-Msh3p heterodimer is thought to stabilize heteroduplex intermediates that form during the repair of DNA plasmids that contain nonhomologous DNA sequences that must be excised prior to initiating DNA synthesis-mediated repair steps (SUGAWARA et al. 1997 ; STUDAMIRE et al. 1999 ). Other MMR factors such as Msh6p or the MutLp homologs are not required in this pathway, suggesting that these msh2 alleles are defective in repair steps following mismatch recognition. msh2-L560S strains were also functional in the DSBR assay, displaying a recombination proficiency that was identical to wild type at the semipermissive temperature of 30° (E. ALANI, N. SUGAWARA and J. E. HABER, data not shown). Functionality in the DSBR assay suggests the region around L-560 of Msh2p is involved in steps following mismatch recognition and/or steps specific to Msh2p-Msh6p-dependent MMR. A defect in activities mediated through Msh2p-Exo1p would also be consistent with this idea. However, we do not believe that this is due to a global defect in Msh2p-Exo1p interactions as the msh2-L560Sp behaved like Msh2p in co-immunoprecipitation studies involving Exo1p (Fig 3).

Synthetic lethality of pol3-01 with MMR-defective mutations:
How can the synthetic lethality between pol3-01 and MMR mutations be explained? MORRISON et al. 1993 proposed that this lethality results from an extremely high mutation rate that leads to nucleotide misincorporation events in essential genes. According to this model, the observed viability of diploids homozygous for both mutations is due to the diploid copy number of essential genes, allowing genetic recombination to repair DNA lesions through an unmutagenized donor sequence. Several studies appear to be inconsistent with the hypothesis that synthetic lethality results from the accumulation of nucleotide misincorporations. First, mutations in other replication factors, including RFC1 and POL2, display synergy for mutation rates in combination with MMR defects in haploid strains without resulting in synthetic lethality (TRAN et al. 1999B ; XIE et al. 1999 ). A caveat in this argument is that these alleles display a less severe mutator phenotype than pol3-01 leaving the possibility of a threshold effect.

Second, the data presented in this report argue against a simple correlation between synthetic lethality and mutation rate as the mutator defect in the msh2^ts alleles in the absence of pol3-01 did not correlate with synthetic lethality. The most extreme case was observed in the canavanine assay, where most of the alleles exhibited a mutation rate that was similar to an msh2 mutant at both the permissive and nonpermissive temperatures for msh2^ts pol3-01 synthetic lethality of these alleles (Table 2). In contrast, many but not all of these same alleles exhibited temperature sensitivity in the lys2-Bgl reversion assay (Table 1). Sequencing of DNA from canavanine-resistant and Lys⁺ papillations from msh2 strains (MARSISCHKY et al. 1996 ; FLORES-ROZAS and KOLODNER 1998 ) revealed that Lys⁺ reversion resulted almost exclusively from single-nucleotide deletions in short mono-nucleotide repeats and canavanine resistance resulted from a mix of nucleotide misincorporations and single-nucleotide deletions. While this difference in mutational spectra was not confirmed for the msh2^ts alleles in this study, the strong phenotype of certain alleles in the canavanine assay at the permissive temperature and the lack of direct correlation to the lys2-Bgl assay suggest that synthetic lethality with pol3-01 is not due to a catastrophic accumulation of these types of mutations and that synthetic lethality results from another type of DNA lesion not detected in these assays (see below).

We hypothesize that synthetic lethality in MMR-defective pol3-01 strains is caused by defects in Pol exonuclease that result in another type of DNA lesion that can be recognized by the Msh2p-Msh6p repair pathway. Because pol3-01 rad52 mutants are viable, we hypothesize that the pol3-01 mutation is unlikely to result in the formation of double-strand breaks that must be corrected by a recombinational repair pathway (XIE et al. 1999 ). Alternatively, on the basis of synthetic lethality of pol3-01 rad27 haploid and diploid strains, KOKOSKA et al. 1998 suggested that the exonuclease activity of Pol plays a critical role in completing Okazaki fragment processing in the absence of Rad27p flap endonuclease activity either by directly removing the final 5' ribonucleotide on the downstream RNA primer or by causing the polymerase to back up, allowing other nuclease activities to complete processing. It is possible that even in a wild-type strain a certain percentage of downstream RNA primers are not completely removed prior to completion of the upstream DNA fragment.

As an extension of the KOKOSKA et al. 1998 model, the proofreading function of Pol may respond to an incompletely processed RNA primer on the downstream DNA fragment and signal polymerase activity mediated through the exonuclease domain to pause or back up, allowing for completion of processing. In the absence of this activity, disassociation of the polymerase could result in short gaps forming between DNA fragments. One possibility is that such pol3-01-induced lesions might be recognized by the Msh2p-Msh6p heterodimer, triggering MMR. The activity of MMR, including Exo1p nuclease, would then excise the nascent DNA fragments and the gap would be filled during the repair process. At present, a gap-binding activity has not been demonstrated for the Msh2p-Msh6p complex but is consistent with the DNA-binding properties of Msh proteins. The Msh2p-Msh6p heterodimer binds specifically to several different types of DNA substrates, including single-nucleotide insertions, palindromic insertions, and synthetic Holliday junctions, and recent chromatin immunoprecipitation experiments suggest that Msh2p can load onto double-strand break sites at recessed ends during genetic recombination (ALANI 1996 ; MARSISCHKY and KOLODNER 1999 ; E. EVANS, N. SUGAWARA, J. HABER and E. ALANI, unpublished results). While there is no direct evidence for pol3-01 induction of short-gapped DNA lesions, it should be noted that in vitro assays did not reveal a significant defect in pol3-01p for DNA polymerization. This assay, however, would not detect other defects that could affect replication processivity and fidelity in vivo, such as a defect in interaction with other replication factors (SIMON et al. 1991 ). Further support of this model will require additional studies of defects in the pol3-01p mutant protein and an analysis of the DNA-binding specificities of the Msh2p-Msh6p and msh2^tsp-Msh6p complexes.

In yeast, DNA damage results in the arrest of cells at one of several cell cycle control checkpoints (LONGHESE et al. 1998 ; WEINERT 1998 ). Strains bearing rad9 mutations are defective in G₂/M cell cycle arrest and die upon induction of DNA damage, presumably due to a failure to repair DNA lesions prior to completion of mitosis (WEINERT and HARTWELL 1988 ). Another way to explain the msh2 pol3-01 lethality is that DNA damage in these strains causes the irreversible activation of a cell cycle checkpoint. Alternatively, DNA lesions in msh2 pol3-01 strains escape detection by the cell cycle checkpoint machinery and cause lethality during mitosis. We have not been able to provide evidence for such models as a rad9 mutation in msh2 pol3-01 strains carrying the six msh2^ts alleles did not confer increased lethality at 26° or increased viability at 35° (data not shown).

Many of the factors involved in initiation of MMR have been identified (reviewed in KOLODNER and MARSISCHKY 1999 ). While several factors have been implicated in downstream repair steps, full definition of the components has not been achieved. Assays to identify new mutator alleles have been only moderately successful (NI et al. 1999 ; TRAN et al. 1999A ; XIE et al. 1999 ). This may be due to the fact that many of these downstream factors may be encoded by essential genes and/or have redundant activities. Suppression of msh2-L560S pol3-01 and msh2-L910P pol3-01 synthetic lethality by overexpression of EXO1 and MSH6, respectively, highlights the power of a suppression approach to identify MMR components. Studies are under way to identify novel MMR factors from a library of yeast genes via dosage suppression of the msh2^ts pol3-01 conditional lethality phenotype.

	FOOTNOTES

¹ Present address: Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139.

	ACKNOWLEDGMENTS

We thank Dan Smith for performing the tetrad dissections described in this study. We also thank Jayson Bowers, Elizabeth Evans, and Nancy Kleckner for helpful discussions and technical advice and Elizabeth Evans and members of the Alani lab for their insightful comments on the manuscript. We are grateful to Gray Crouse, Connie Holm, Richard Kolodner, Robert Lahue, Tom Petes, and Louise Prakash for plasmids and strains. T.S. was supported by a National Institutes of Health predoctoral training grant. E.A. was supported by National Institutes of Health grant R01-GM53085 and U.S. Department of Agriculture Hatch grant NYC-1656424.

Manuscript received January 10, 2000; Accepted for publication February 21, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

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].

ALANI, E., T. SOKOLSKY, B. STUDAMIRE, J. J. MIRET, and R. S. LAHUE, 1997  Genetic and biochemical analysis of Msh2p-Msh6p: role of ATP hydrolysis and Msh2p-Msh6p subunit interactions in mismatch base pair recognition. Mol. Cell. Biol. 17:2436-2447[Abstract].

BAMBARA, R. A., R. S. MURANTE, and L. A. HENRICKSEN, 1997  Enzymes and reactions at the eukaryotic DNA replication fork. J. Biol. Chem. 272:4647-4650[Free Full Text].

BOWERS, J., T. SOKOLSKY, T. QUACH, and E. ALANI, 1999  A mutation in the MSH6 subunit of the Saccharomyces cerevisiae MSH2-MSH6 complex disrupts mismatch recognition. J. Biol. Chem. 274:16115-16125[Abstract/Free Full Text].

BRADFORD, M. M., 1976  A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254[Medline].

CHEN, C., B. J. MERRILL, P. J. LAU, C. HOLM, and R. D. KOLODNER, 1999  Saccharomyces cerevisiae pol30 (proliferating cell nuclear antigen) mutations impair replication fidelity and mismatch repair. Mol. Cell. Biol. 19:7801-7815[Abstract/Free Full Text].

CHRISTIANSON, T. W., R. S. SIKORSKI, M. DANTE, J. H. SHERO, and P. HIETER, 1992  Multifunctional yeast high-copy-number shuttle vectors. Gene 110:119-122[Medline].

CROUSE, G. F., 1998 Mismatch repair systems in Saccharomyces cerevisiae, pp. 411–448 in DNA Damage and Repair—Biochemistry, Genetics and Cell Biology, edited by J. NICKOLOFF and M. HOEKSTRA. Humana Press, Clifton, NJ.

FIORENTINI, P., K. N. HUANG, D. X. TISHKOFF, R. D. KOLODNER, and L. S. SYMINGTON, 1997  Exonuclease I of Saccharomyces cerevisiae functions in mitotic recombination in vivo and in vitro. Mol. Cell. Biol. 17:2764-2773[Abstract].

FLORES-ROZAS, H. and R. D. KOLODNER, 1998  The Saccharomyces cerevisiae MLH3 gene functions in MSH3-dependent suppression of frameshift mutations. Proc. Natl. Acad. Sci. USA 95:12404-12409[Abstract/Free Full Text].

GARY, R., M. S. PARK, J. P. NOLAN, H. L. CORNELIUS, and O. G. KOZYREVA et al., 1999  A novel role in DNA metabolism for the binding of Fen1/Rad27 to PCNA and implications for genetic risk. Mol. Cell. Biol. 19:5373-5382[Abstract/Free Full Text].

GIETZ, R. D. and R. H. SCHIESTL, 1991  Applications of high efficiency lithium acetate transformation of intact yeast cells using single-stranded nucleic acids as carrier. Yeast 7:253-263[Medline].

GREENE, C. N. and S. JINKS-ROBERTSON, 1997  Frameshift intermediates in homopolymer runs are removed efficiently by yeast mismatch repair proteins. Mol. Cell. Biol. 17:2844-2850[Abstract].

GU, L., Y. HONG, S. MCCULLOCH, H. WATANABE, and G. M. LI, 1998  ATP-dependent interaction of human mismatch repair proteins and dual role of PCNA in mismatch repair. Nucleic Acids Res. 26:1173-1178[Abstract/Free Full Text].

GUERRETTE, S., T. WILSON, S. GRADIA, and R. FISHEL, 1998  Interactions of human hMSH2 with hMSH3 and hMSH2 with hMSH6: examination of mutations found in hereditary nonpolyposis colorectal cancer. Mol. Cell. Biol. 18:6616-6623[Abstract/Free Full Text].

HO, S. N., H. D. HUNT, R. M. HORTON, J. K. PULLEN, and L. R. PEASE, 1989  Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51-59[Medline].

JOHNSON, R. E., G. K. KOVVALI, S. N. GUZDER, N. S. AMIN, and C. HOLM et al., 1996  Evidence for involvement of yeast proliferating cell nuclear antigen in DNA mismatch repair. J. Biol. Chem. 271:27987-27990[Abstract/Free Full Text].

KESTI, T. and J. E. SYVAOJA, 1991  Identification and tryptic cleavage of the catalytic core of HeLa and calf thymus DNA polymerase epsilon. J. Biol. Chem. 266:6336-6341[Abstract/Free Full Text].

KOKOSKA, R. J., L. STEFANOVIC, H. T. TRAN, M. A. RESNICK, and D. A. GORDENIN et al., 1998  Destabilization of yeast micro- and minisatellite DNA sequences by mutations affecting a nuclease involved in Okazaki fragment processing (rad27) and DNA polymerase delta (pol3-t). Mol. Cell. Biol. 18:2779-2788[Abstract/Free Full Text].

KOLODNER, R. D. and G. T. MARSISCHKY, 1999  Eukaryotic DNA mismatch repair. Curr. Opin. Genet. Dev. 9:89-96[Medline].

LEA, D. E. and C. A. COULSON, 1949  The distribution of the numbers of mutants in bacterial populations. J. Genet. 49:264-285.

LIEBER, M. R., 1997  The FEN-1 family of structure-specific nucleases in eukaryotic DNA replication, recombination and repair. Bioessays 19:233-240[Medline].

LONGHESE, M. P., M. FOIANI, M. MUZI-FALCONI, G. LUCCHINI, and P. PLEVANI, 1998  DNA damage checkpoint in budding yeast. EMBO J. 17:5525-5528[Medline].

MANIATIS, T., E. F. FRITSCH and J. SAMBROOK, 1982 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MARSISCHKY, G. T. and R. D. KOLODNER, 1999  Biochemical characterization of the interaction between the Saccharomyces cerevisiae MSH2-MSH6 complex and mispaired bases in DNA. J. Biol. Chem. 274:26668-26682[Abstract/Free Full Text].

MARSISCHKY, G. T., N. FILOSI, M. F. KANE, and R. KOLODNER, 1996  Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 in MSH2-dependent mismatch repair. Genes Dev. 10:407-420[Abstract/Free Full Text].

MCALEAR, M. A., K. M. TUFFO, and C. HOLM, 1996  The large subunit of replication factor C (Rfc1p/Cdc44p) is required for DNA replication and DNA repair in Saccharomyces cerevisiae.. Genetics 142:65-78[Abstract].

MODRICH, P., 1997  Strand-specific mismatch repair in mammalian cells. J. Biol. Chem. 272:24727-24730[Free Full Text].

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

MORRISON, A. and A. SUGINO, 1994  The 3'5' exonucleases of both DNA polymerases delta and epsilon participate in correcting errors of DNA replication in Saccharomyces cerevisiae.. Mol. Gen. Genet. 242:289-296[Medline].

MORRISON, A., J. B. BELL, T. A. KUNKEL, and A. SUGINO, 1991  Eukaryotic DNA polymerase amino acid sequence required for 3'-5' exonuclease activity. Proc. Natl. Acad. Sci. USA 88:9473-9477[Abstract/Free Full Text].

MORRISON, A., A. L. JOHNSON, L. H. JOHNSTON, and A. SUGINO, 1993  Pathway correcting DNA replication errors in Saccharomyces cerevisiae.. EMBO J. 12:1467-1473[Medline].

NI, T. T., G. T. MARSISCHKY, and R. D. KOLODNER, 1999  MSH2 and MSH6 are required for removal of adenine misincorporated opposite 8-oxo-guanine in S. cerevisiae.. Mol. Cell 4:439-444[Medline].

QIU, J., M. X. GUAN, A. M. BAILIS, and B. SHEN, 1998  Saccharomyces cerevisiae exonuclease-1 plays a role in UV resistance that is distinct from nucleotide excision repair. Nucleic Acids Res. 26:3077-3083[Abstract/Free Full Text].

QIU, J., Y. QIAN, V. CHEN, M. X. GUAN, and B. SHEN, 1999  Human exonuclease 1 functionally complements its yeast homologues in DNA recombination, RNA primer removal, and mutation avoidance. J. Biol. Chem. 274:17893-17900[Abstract/Free Full Text].

REENAN, R. A. 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].

ROSE, M. D., F. WINSTON and P. HIETER, 1990 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

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].

SHEN, B., J. P. NOLAN, L. A. SKLAR, and M. S. PARK, 1996  Essential amino acids for substrate binding and catalysis of human flap endonuclease 1. J. Biol. Chem. 271:9173-9176[Abstract/Free Full Text].

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

SIMON, M., L. GIOT, and G. FAYE, 1991  The 3' to 5' exonuclease activity located in the DNA polymerase delta subunit of Saccharomyces cerevisiae is required for accurate replication. EMBO J. 10:2165-2170[Medline].

SOMMERS, C. H., E. J. MILLER, B. DUJON, S. PRAKASH, and L. PRAKASH, 1995  Conditional lethality of null mutations in RTH1 that encodes the yeast counterpart of a mammalian 5'- to 3'-exonuclease required for lagging strand DNA synthesis in reconstituted systems. J. Biol. Chem. 270:4193-4196[Abstract/Free Full Text].

STRAND, M., T. A. PROLLA, R. M. LISKAY, and T. D. PETES, 1993  Destabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature 365:274-276[Medline].

STREISINGER, G., Y. OKADA, J. EMRICH, J. NEWTON, and A. TSUGITA et al., 1966  Frameshift mutations and the genetic code. Cold Spring Harb. Symp. Quant. Biol. 31:77-84[Medline].

STUDAMIRE, B., G. PRICE, N. SUGAWARA, J. E. HABER, and E. ALANI, 1999  Separation-of-function mutations in Saccharomyces cerevisiae MSH2 that confer mismatch repair defects but do not affect nonhomologous-tail removal during recombination. Mol. Cell. Biol. 19:7558-7567[Abstract/Free Full Text].

SUGAWARA, N., F. PAQUES, M. COLAIACOVO, and J. E. HABER, 1997  Role of Saccharomyces cerevisiae Msh2 and Msh3 repair proteins in double-strand break-induced recombination. Proc. Natl. Acad. Sci. USA 94:9214-9219[Abstract/Free Full Text].

SZANKASI, P. and G. R. SMITH, 1995  A role for exonuclease I from S. pombe in mutation avoidance and mismatch correction. Science 267:1166-1169[Abstract/Free Full Text].

TISHKOFF, D. X., A. L. BOERGER, P. B