Control of GT Repeat Stability in Schizosaccharomyces pombe by Mismatch Repair Factors

Corresponding author: Oliver Fleck, Institute of Cell Biology, University of Bern, Baltzer-Strasse 4, CH-3012 Bern, Switzerland., fleck{at}izb.unibe.ch (E-mail)

Communicating editor: M. LICHTEN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The mismatch repair (MMR) system ensures genome integrity by removing mispaired and unpaired bases that originate during replication. A major source of mutational changes is strand slippage in repetitive DNA sequences without concomitant repair. We established a genetic assay that allows measuring the stability of GT repeats in the ade6 gene of Schizosaccharomyces pombe. In repair-proficient strains most of the repeat variations were insertions, with addition of two nucleotides being the most frequent event. GT repeats were highly destabilized in strains defective in msh2 or pms1. In these backgrounds, mainly 2-bp insertions and 2-bp deletions occurred. Surprisingly, essentially the same high mutation rate was found with mutants defective in msh6. In contrast, a defect in swi4 (a homologue of Msh3) caused only slight effects, and instability was not further increased in msh6 swi4 double mutants. Also inactivation of exo1, which encodes an exonuclease that has an MMR-dependent function in repair of base-base mismatches, caused only slightly increased repeat instability. We conclude that Msh2, Msh6, and Pms1 have an important role in preventing tract length variations in dinucleotide repeats. Exo1 and Swi4 have a minor function, which is at least partially independent of MMR.

MISMATCHED and unpaired bases arise during replication by misincorporation of nucleotides and strand slippage, respectively. The major system directed to repair base-base mismatches and loops with one or a few unpaired nucleotides is the bacterial mutHLS pathway and related but more complex systems in eukaryotes (MODRICH 1991 ; JIRICNY 1998 , JIRICNY 2000 ; KOLODNER and MARSISCHKY 1999 ; MARRA and SCHAR 1999 ). In eukaryotes, three MutS homologues, Msh2, Msh3, and Msh6, and three MutL homologues, termed Pms2, Mlh1, and Mlh3 in human (Pms1, Mlh1, and Mlh3 in Saccharomyces cerevisiae), are thought to be components of the nuclear mismatch repair (MMR) system.

Two heterodimers, Msh2-Msh6 (MutS) and Msh2-Msh3 (MutSß), are involved in the recognition step. After binding of either MutS or MutSß to the DNA heterology, a MutL heterodimer is recruited to the MutS-DNA complex. Mlh1-Pms2 is involved in MutS- and MutSß-mediated repair of mismatches and loops (UMAR and KUNKEL 1996 ; BUERMEYER et al. 1999 ; KOLODNER and MARSISCHKY 1999 ; MARRA and SCHAR 1999 ; JIRICNY 2000 ). Mlh1-Mlh3 plays a minor role in repair of loops bound by MutSß and might rather have its main function in recombination (FLORES-ROZAS and KOLODNER 1998 ; NAKAGAWA et al. 1999 ; WANG et al. 1999 ).

Mutations in the human MMR genes MSH2 and MLH1 cause predisposition to hereditary nonpolyposis colon cancer and other types of cancer. In contrast, mutations in PMS2, MSH3, and MSH6 are rarely found to be correlated with cancer (UMAR and KUNKEL 1996 ; BUERMEYER et al. 1999 ). The role of MLH3 in cancer development remains to be analyzed in more detail, as this gene was just recently discovered (LIPKIN et al. 2000 ). A defect in MSH2, PMS2, or MLH1 results in increased mutation rates and high instability of microsatellites (UMAR and KUNKEL 1996 ; BUERMEYER et al. 1999 ). Microsatellites are repetitive sequences consisting of one to several nucleotides per repeat unit and are common motifs in genomes of higher eukaryotes. Instability of repetitive sequences is thought to be caused by frequent strand slippage during replication without subsequent repair. When the slipped and unpaired bases are in the newly synthesized strand, additions of repeat units are the consequence, whereas slippage of the template strand results in deletions.

Microsatellite stability in S. cerevisiae is also maintained by the MMR system with Msh2, Pms1, and Mlh1 as key players (STRAND et al. 1993 ; JOHNSON et al. 1996 ; GREENE and JINKS-ROBERTSON 1997 ; SIA et al. 1997 ). A defect in any of the respective genes causes greatly increased rates of length variations. Loss of Msh6 has little effect on repeats with one or two nucleotides per unit. Such repeats are moderately affected in msh3 mutants; i.e., tract alterations occur more often than in msh6 mutants but clearly not as frequently as in msh2 mutants. Nevertheless, a msh3 msh6 double mutant shows essentially the same high instability as the msh2 mutant. The genetic data originally led to the conception that MutS and MutSß can redundantly repair small loops with one or two unpaired nucleotides. However, current studies on human MMR indicate that MutS is the major recognition factor for base-base mismatches and small loops while MutSß rather serves as a backup system (JIRICNY 1998 ; KOLODNER and MARSISCHKY 1999 ). This idea is based on studies with human MutS and MutSß purified from HeLa cells and is strengthened by the observations that the cellular level of MutS is much higher than that of MutSß (DRUMMOND et al. 1997 ; GENSCHEL et al. 1998 ; MARRA et al. 1998 ). Furthermore, microsatellite instability is not increased in MSH3-deficient human cell lines, while microsatellites of mononucleotide runs are affected in MSH6-mutated cell lines (INOKUCHI et al. 1995 ; PAPADOPOULOS et al. 1995 ).

Not much is known about proteins acting further downstream in the MMR system. One factor is Exo1, a 5' to 3' exonuclease, which was identified in Schizosaccharomyces pombe, S. cerevisiae, mouse, and human (SZANKASI and SMITH 1995 ; TISHKOFF et al. 1997 , TISHKOFF et al. 1998 ; SCHMUTTE et al. 1998 ; WILSON et al. 1998 ). A mutated exo1 causes increased mutation rates in epistatic relationship to msh2 (TISHKOFF et al. 1997 ; RUDOLPH et al. 1998 ). These studies and the ability of Exo1 to interact with Msh2 (TISHKOFF et al. 1997 ) suggest a function in the MMR system. However, it should be noted that Exo1 is also involved in recombinational processes (FIORENTINI et al. 1997 ).

In S. pombe, the MutS homologues Swi4 (most closely related to the Msh3 subgroup), Msh2, and Msh6, and the MutL homologue Pms1 have been identified and characterized (FLECK et al. 1992 ; SCHAR et al. 1997 ; RUDOLPH et al. 1999 ; TORNIER et al. 2001 , accompanying article). Msh2, Msh6, and Pms1 turned out to be important components of MMR in repair of base-base mismatches and one-nucleotide loops, while no significant defects in mismatch repair were found in swi4 mutants. Swi4 has a function in the termination step of mating-type switching, likely in conjunction with Msh2 (FLECK et al. 1992 ; RUDOLPH et al. 1999 ).

The aim of the work presented here was to analyze the role of Msh2, Msh6, Swi4, Pms1, and Exo1 of S. pombe in stability of GT repeats. We constructed strains with insertion of GT repeats in the ade6 marker gene and measured mutation rates in wild type and in mutants defective in one or more of the MMR genes. To obtain information about the relative contribution of MutS and MutSß in repair of dinucleotide loops in S. pombe, we were particularly interested in determining the effects caused by loss of Msh2, Msh6, and Swi4.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

General yeast genetic methods and media:
The S. pombe media YEA (yeast extract agar), YEL (yeast extract liquid), MEA (malt extract agar), and MMA (minimal medium agar) and general genetic methods were used as described elsewhere (GUTZ et al. 1974 ; SCHAR et al. 1997 ).

Construction of S. pombe strains with GT repeats in the ade6 marker gene:
pAN-T, a pUC18 derivative, contains a XhoI-EcoRI fragment of the ade6 gene of S. pombe, in which an internal 340-bp DraIII-HindIII fragment was replaced by the Escherichia coli tetracycline resistance gene. The tetracycline resistance gene was then replaced by the previously deleted DraIII-HindIII fragment derived from ade6, but additionally containing the (GT)₈ repeat close to the DraIII site. This fragment was obtained by restriction digestion of a PCR product obtained with primers ade6-GT8, 5'-AAGCACTTGGTGATCGTGTGTGTGTGTGTGTCCGCTTTATGTTGAAAAAGTT-3' and ade6-H, 5'-GGGCAAGCTTCAATGGTGTA-3' and the ade6-containing vector pCG162 (GRIMM et al. 1994 ) as template. Plasmids were isolated from E. coli transformants, which lost resistance to tetracycline and were checked for the presence of the desired insert by restriction digestions and DNA sequencing.

The (GT)₈ repeat was then introduced into the S. pombe ade6 gene by homologous recombination of the 1.7-kb XhoI-EcoRI fragment transformed into the S. pombe strain AM1 (h^- ade6::ura4⁺ his3-D1 leu1-32 ura4-D18), giving rise to AM3 (h^- ade6-[(GT)₈-1397] his3-D1 leu1-32 ura4-D18). AM1 is a derivative of strain PS5, in which the 0.8-kb BamHI-HindIII fragment of ade6 is replaced by the ura4⁺ gene (SCHUCHERT et al. 1991 ). Transformants were grown on fluoroorotic acid plates to select for loss of the ura4⁺ marker as described previously (GRIMM et al. 1988 ). Correct integration was then proved by PCR and sequencing (see below). All strains with a (GT)₈ repeat were obtained from crosses with AM3. During fluctuation tests, the strain OL539 (h^- msh2::his3⁺ ade6-[(GT)₁₀-1397] his3-D1 ura4-D18) containing a (GT)₁₀ repeat and the strain OL540 (h^- msh2::his3⁺ ade6-[(GT)₉-1397] his3-D1 ura4-D18) with a (GT)₉ repeat were isolated and identified by sequencing. The repeats were combined by crosses with the various marker-disrupted repair genes, which were studied in this work: msh2::his3⁺ (RUDOLPH et al. 1999 ), msh6::arg3⁺ (TORNIER et al. 2001 , accompanying article), swi4::ura4⁺ (FLECK et al. 1992 ), pms1::ura4⁺ (SCHAR et al. 1997 ), pms1::his3⁺ (strain collection of Bern), and exo1::ura4⁺ (SZANKASI and SMITH 1995 ; RUDOLPH et al. 1998 ).

Determination of GT repeat instability:
All repeats used in this study are insertions at an existing GT dinucleotide in the ade6 gene at position 1397, with the ATG start codon at position 875 and the stop codon at position 2531 according to the nomenclature of SZANKASI et al. 1988 . (GT)₈ and (GT)₉ (insertions of 14 and 16 bp, respectively) produce in frame stop codons located 47 and 12 bp downstream of the insertion, respectively, and cause adenine auxotrophy. In contrast, (GT)₁₀ represents an in frame insertion of 18 bp, allowing growth on minimal medium without adenine. Instability of GT repeats was measured by fluctuation tests as described previously (SCHAR et al. 1997 ; RUDOLPH et al. 1998 , RUDOLPH et al. 1999 ). In the case of strains with (GT)₈ or (GT)₉ repeats, seven small colonies were each inoculated in 5-ml YEL and grown to stationary phase. Appropriate amounts were plated on minimal medium for selection of Ade⁺ revertants and on minimal medium supplemented with adenine for determination of cell titers. A maximum of 5 x 10⁷ cells were streaked out per plate, a density that does not significantly affect growth of Ade⁺ revertants (data not shown). After 7 days of growth at 30°, mutation rates were calculated from the median number of Ade⁺ revertants per total cell number of the culture (LEA and COULSON 1949 ). In the case of (GT)₁₀, one small colony was picked from a minimal medium plate, suspended in 0.5 ml 0.85% NaCl, and diluted 1:100. About 15 1-ml YEL tubes were each inoculated with 100 cells and grown for 24–36 hr until 10,000–30,000 cells were present per culture. The total of the cultures was plated out on YEA, allowing growth of all cells. After 4 days of growth at 30°, red colonies (for colony color see below) were counted and tested for adenine auxotrophy. Mutation rates were calculated either from the median or by the zero method from the number of cultures without Ade^- colonies per total number of cultures.

Determination of tract length variations of GT repeats:
Variations in the number of GT dinucleotide units were determined by visual inspection of the colony color and by DNA sequencing. ade6 mutants form red colonies on YEA with a limited amount of adenine and cannot grow on minimal medium without adenine. Strains with a (GT)₈ or (GT)₉ repeat in ade6 form red colonies. Among the Ade⁺ revertants, (GT)₄, (GT)₇, and (GT)₁₀ repeats were identified by sequencing (Table 1), which all restored the reading frame of ade6. No other events were detected. Strains with (GT)₄ or (GT)₇ repeats form white colonies on YEA and MMA plates, like ade6⁺ wild-type strains. In contrast, strains containing a (GT)₁₀ repeat form pink colonies on both types of medium, indicating that the activity of the ade6 gene product is not completely retained. Thus the simple determination of the colony color of Ade⁺ revertants derived from strains with either a (GT)₈ or (GT)₉ repeat allows discrimination between deletions [(GT)₄ or (GT)₇] and insertions [(GT)₁₀] of repeat units (Fig 1). The mutational spectra were determined from the sum of pink and of white colonies of all cultures examined.

View larger version (19K): In this window In a new window Download PPT slide   Figure 1. Test system for GT repeat stability. (A) GT repeats were introduced at position 1397 of the ade6 gene (the ATG start codon is at position 875). (GT)8 is shown as an example (underlined) and was created by insertion of 7 GT units (letters in lowercase) at an existing GT site. The insertion causes a frame shift and an in frame stop codon downstream. (B) Mutation rates and spectra detectable with the repeats (GT)8, (GT)9, and (GT)10. (GT)8- and (GT)9-containing strains are Ade- and form red colonies on YEA (solid circles). Ade+ revertants occur either by deletions or insertions of repeat units (changes of the nucleotide numbers are indicated). Strains with (GT)7 and (GT)4 form white colonies (open circle), while (GT)10 strains form pink colonies (shaded circles). With the (GT)10 repeat events are detectable that lead to Ade- (solid circles).

DNA sequences were determined by direct sequencing of PCR products (DORIT et al. 1995 ). PCR was performed on chromosomal DNA using primers ade6-1275, 5'-GGCCAAGAGTTTGGTTATCC-3' and ade6-H, 5'-GGGCAAGCTTCAATGGTGTA-3' under standard conditions (5 min at 94°; 30 cycles of 45 sec at 94°, 45 sec at 50°, 45 sec at 72°; 10 min at 72°). The PCR product was purified from primers by precipitation with 2.5 M ammonium acetate/1 volume absolute ethanol and used for amplification of a single strand with primer ade6-H (5 min at 94°; 20 cycles of 45 sec at 94°, 45 sec at 48°, 45 sec at 72°; 10 min at 72°). After removal of free primers, sequences were determined with primer ade6-1275 by the dideoxy method using a sequencing kit (Amersham Pharmacia, Dübendorf, Switzerland).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Test system to measure GT repeat instability in S. pombe:
In this work we were interested in studying the instability of GT dinucleotide repeats in S. pombe. The emphasis was to understand the role of genes with a known function in repair of base-base mismatches, i.e., msh2, pms1, msh6, and exo1. We also analyzed swi4, which has no detectable function in base-base mismatch repair but might have a function in loop repair, similar to the situation in S. cerevisiae (JOHNSON et al. 1996 ; GREENE and JINKS-ROBERTSON 1997 ; SIA et al. 1997 ). The assays to measure GT repeat instability are illustrated in Fig 1. We introduced GT repeats at an existing GT dinucleotide into the ade6 gene as described in MATERIALS AND METHODS. A region inside the ade6 gene was chosen, where none of the known mutations is located, in the hope that in frame additions of several nucleotides do not disrupt the function of the ade6 gene. We tested strains in which seven, eight, or nine GT repeat units were inserted at an existing GT site and thus tested instability of (GT)₈, (GT)₉, and (GT)₁₀ repeats.

(GT)₈ and (GT)₉ repeats produce frame shifts of +2 and +1 nucleotide, respectively. Strains with (GT)₈ and (GT)₉ are Ade^- and form red colonies on media with a limited amount of adenine (MATERIALS AND METHODS). Strains with either (GT)₈ or (GT)₉ were used to measure reversions to Ade⁺; +4-, -2-, and -8-bp changes are detectable with (GT)₈ strains, and +2-, -4-, and -10-bp changes with (GT)₉ strains. These changes resulted in (GT)₁₀, (GT)₇, and (GT)₄ repeats, respectively (Fig 1). During the fluctuation tests, we noticed that Ade⁺ revertants formed either white or pink colonies. DNA sequencing of 56 pink colonies derived from the various strain backgrounds exclusively revealed (GT)₁₀ repeats and sequencing of 60 white colonies, either (GT)₇ or (GT)₄ repeats (Table 1). No Ade⁺ revertants with other variations in repeat tract length were found. Thus, starting with ade6-mutated strains containing either a (GT)₈ or (GT)₉ repeat, Ade⁺ revertants with additions of repeat units can be easily distinguished by their colony color from those with deletions (Fig 1; Table 1). Moreover, the use of (GT)₈ and (GT)₉ strains allows determination of how frequently specific events occur in the various strain backgrounds. Due to the possibility of selection for Ade⁺ revertants, even relatively rare events and small effects can be detected. However, it should be kept in mind that not all events are detectable when either (GT)₈ or (GT)₉ were assayed. Therefore, tract length variations in a (GT)₁₀ repeat were also analyzed, where most of the possible changes are detectable. The (GT)₁₀ repeat includes 18 additional nucleotides and thus an in frame insertion. Strains containing a (GT)₁₀ repeat were used to measure all events (+4, +2, -2, -4 bp, and others) leading to red colonies (Ade^-).

Requirement for mismatch repair genes in stability of GT dinucleotide repeats:
ade6 mutants containing either a (GT)₈ or (GT)₉ repeat were used to measure reversion rates to Ade⁺ by fluctuation tests as described in MATERIALS AND METHODS. In mismatch repair-proficient wild type, reversions occurred rarely in the case of the (GT)₈ repeat, but quite frequently for (GT)₉, with a rate 160 times higher than that for (GT)₈ (Table 2). The reversion rates were dramatically increased when msh2, msh6, or pms1 were defective. We found rates of 5 x 10^-5 for (GT)₈, and of 1.7 x 10^-4 for (GT)₉, which correspond to 15,000- and 300-fold increases, respectively. In contrast, inactivation of swi4 caused only a 12-fold increase of reversions of (GT)₈ and, surprisingly, a 3-fold decrease of (GT)₉ reversions. msh2 swi4 and msh6 swi4 double mutants showed the same high reversion rates as msh2 and msh6 single mutants in the case of (GT)₈, but an 4-fold reduction in the case of (GT)₉, when compared with msh2 and msh6. In addition, the triple mutant msh2 msh6 swi4 also showed a 10 times lower (GT)₉ reversion rate than msh2 msh6 (Table 2).

When exo1 was mutated, a nine-fold increase was found for (GT)₈ and a two-fold increase for (GT)₉ (Table 2). Reversion of (GT)₈ in the swi4 exo1 double mutant was in the same range as in respective single mutants. In contrast, reversion rates of (GT)₉ were reduced, which is similar to swi4 and different from exo1. The msh2 exo1 and pms1 exo1 mutants showed extreme instability of both (GT)₈ and (GT)₉. We measured reversion rates in the range of 5 x 10^-4, which is significantly higher than the rates caused by inactivation of either msh2 or pms1 alone.

The nature of the reversions in the various strain backgrounds was determined by DNA sequencing and by inspection of the colony color. Sequencing of revertants with pink colonies exclusively revealed a (GT)₁₀ repeat and of revertants with white colonies either a (GT)₇ or (GT)₄ repeat (Table 1). The distributions of deletions and insertions in the (GT)₈ and the (GT)₉ repeats are summarized in Table 3. In all strain backgrounds reversion of (GT)₉ occurred in almost all of the cases (93–100%) by insertion of two nucleotides. In wild type, 83% of Ade⁺ revertants of (GT)₈ repeats produced pink colonies. Thus, (GT)₈ reverted mainly by insertion of four nucleotides. Among the 17% Ade⁺ with white colonies, both 2-bp and 8-bp deletions were detected (Table 1). No big difference in the pattern was observed for exo1 mutants. A total of 73% of (GT)₈ revertants and 99% of (GT)₉ revertants formed pink colonies and thus derived from insertion of 4 and 2 bp, respectively, both resulting in (GT)₁₀ repeats. Among the white (GT)₈ revertants sequenced, 10 with a (GT)₇ repeat and 1 with a (GT)₄ repeat were identified (Table 1).

When either msh2, msh6, or pms1 was mutated, almost all of the (GT)₈ revertants formed white colonies (Table 3). All white revertants from msh2, msh6, or pms1 that were analyzed by sequencing contained a (GT)₇ repeat (Table 1). Most of the (GT)₉ revertants formed pink colonies and thus were caused by insertion of two nucleotides. Pink colonies derived from (GT)₈ and white colonies derived from (GT)₉ occurred with low frequencies (Table 3). Sequencing of white revertants derived from (GT)₉ revealed only (GT)₇ repeats (Table 1). We conclude that in msh2, msh6, and pms1 genetic backgrounds most (GT)₈ reversions were due to deletions of two nucleotides, and most (GT)₉ reversions were due to insertions of two nucleotides. Insertion or deletion of four nucleotides was relatively rare (Table 3), and larger deletions were not detected (Table 1). Insertions of more than four nucleotides probably do not give a functional ade6 gene and therefore cannot be detected with the (GT)₈ or (GT)₉ assays.

In swi4 background, 78% of (GT)₈ revertants showed white colonies (Table 3). Sequencing of six white (GT)₈ revertants revealed that all contained (GT)₇ (Table 1). The rate of 4-bp insertions was only slightly increased and events detectable with the (GT)₉ assay (+2, -4, -10) seem to be rather reduced (Table 2). Thus, the majority of events caused by a defect in swi4 were deletions of two nucleotides. The distributions of msh2 swi4, msh6 swi4, msh2 exo1, and pms1 exo1 double mutants were similar to those of strains mutated in msh2, msh6, or pms1 (Table 3), obviously due to the high mutation rates caused by these mutations in comparison to the weak effects caused by swi4 or exo1. In the swi4 exo1 double mutant, 69% of the reversion events in the (GT)₈ repeat were deletions (Table 3). This pattern is different from that of the respective single mutants, but more similar to that of swi4.

Our analyses of GT repeat instability were extended by testing the (GT)₁₀ repeat. (GT)₁₀ is an insertion of nine GT units at an existing GT dinucleotide and thus an in frame insertion of 18 bp. The (GT)₁₀ assay can be used to measure all changes in the repeat that lead to frame shifts (Fig 1). Fluctuation tests were performed with wild type and the single mutants msh2, msh6, swi4, and exo1, as well as with the double mutant msh2 exo1 (Table 4). The repair-proficient wild type showed a rate of 1.5 x 10^-6. Among the two Ade^- isolated from 60 cultures, we found one 2 and one 4-bp insertion. In msh2 and msh6 mutants, the mutation rate was increased 60- and 120-fold, respectively. With one exception (a 4-bp insertion in msh6 background), all of the events were either 2-bp insertions or 2-bp deletions, both of which occurred with similar frequencies (Table 5). A 4-fold increase was measured with exo1 (Table 4). The mutation rate of the msh2 exo1 mutant was increased 150-fold, which is higher than that of msh2, although not very different from that measured with msh6 (Table 4). No Ade^- colonies were found among 70 cultures of the swi4 mutant. Thus the mutation rate seems to be even lower than that in wild type, although the data are not significantly different (² = 2.37, ²_0.05 = 3.84). However, a swi4-dependent reduction was also found with the (GT)₉ assay (Table 2).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mainly insertions occurred in GT repeats of wild-type cells:
We have established an assay in S. pombe that allows measurement of the stability of GT repeats, a frequent type of microsatellites in DNA. We tested GT repeats with a length of 8, 9, or 10 units, which enabled us to detect different types of insertions and deletions (Fig 1). In repair-proficient cells the majority of the detectable changes are insertions of two nucleotides (1 GT unit). In addition, 4-nucleotide insertions occur more frequently than 2 deletions (Table 3). The bias toward insertions can be explained by more frequent slippage of the newly synthesized strand and/or less frequent repair of loops in the newly synthesized strand. Since similar frequencies of insertions and deletions were found in msh2 and msh6 mutants (Table 5), it is rather likely that slippage of the template occurs as frequently as slippage of the nascent strand. Thus, loops in the newly synthesized strand quite frequently escape repair in wild-type cells. In wild type, quite low reversion rates were found for the (GT)₈ repeat, in contrast to the (GT)₉ repeat, which reverted 160 times more frequently to Ade⁺. This bias is likely due to the selection for events restoring the open reading frame of ade6⁺. Thus, 2-bp insertions for example, likely occur with similar frequencies in (GT)₈ and (GT)₉ repeats, but are only detectable in the latter case.

Control of GT repeat stability by MMR factors:
A defect in the mismatch repair genes msh2, msh6, or pms1 caused highly increased destabilization of GT repeats (Table 2). The major events when either one of these genes was defective were insertions and deletions of two nucleotides (1 GT unit). Reversions from (GT)₉ mainly occurred by 2-bp insertions, while most of the reversions of (GT)₈ were 2-bp deletions (Table 1 and Table 3). Although the rates were about three to four times higher for (GT)₉, +2 and -2 events were equally produced in the (GT)₁₀ repeat in msh2 and msh6 mutants (Table 5). Thus the increased reversion rates of (GT)₉ compared to (GT)₈ were rather due to the presence of 1 additional repeat unit, which might cause more frequent strand slippage. In S. cerevisiae, instability of GT repeats concomitantly increased with increasing tract sizes, both in wild type and in msh2 mutant cells (WIERDL et al. 1997 ).

Minor roles of Exo1 and Swi4:
Loss of exo1 function caused only slightly increased instability of GT repeats, in contrast to msh2, msh6, and pms1. However, mutation rates were further increased in msh2 exo1 and pms1 exo1 double mutants when compared with respective single mutants (Table 2). Thus, exo1 might have a function in GT loop repair, which is independent of the MMR pathway. Our data also suggest that the contribution of exo1 is different in GT repeat stability and in repair of base-base mismatches. In reversion assays of defined point mutations, exo1 mutants showed an increase of mutation rates that is 50% of that of msh2 and pms1 mutants (SZANKASI and SMITH 1995 ; SCHAR et al. 1997 ; RUDOLPH et al. 1998 ). In addition, it was shown that exo1 acts in the msh2-dependent pathway for correction of base-base mismatches (RUDOLPH et al. 1998 ). Epistasis analysis revealed that a defect of exo1 is synergistic with msh2 and pms1 mutations in GT loop repair (Table 2). However, we cannot exclude that exo1 has an additional minor function in the MMR pathway of loop repair.

A defect in swi4 caused only weak effects on GT repeat instability. This finding is in sharp contrast to the situation in S. cerevisiae: GT repeats are more unstable in msh3 mutants than in msh6 mutants, while the high instability observed with msh2 is achieved only in the msh3 msh6 double mutant (JOHNSON et al. 1996 ; SIA et al. 1997 ). These and further studies with S. cerevisiae suggest that small loops are redundantly repaired by the heterodimers Msh2-Msh3 and Msh2-Msh6 (ALANI 1996 ; HABRAKEN et al. 1996 ; IACCARINO et al. 1996 ; MARSISCHKY et al. 1996 ). Our analysis about GT repeat stability in S. pombe showed that Msh6 has the same important function as Msh2, while Swi4 has only a minor and probably different function. The reversion rate of (GT)₈ was not elevated in msh6 swi4 as compared to msh6. Moreover, the mutation rate of (GT)₉ and presumably also of (GT)₁₀ was rather reduced in swi4 mutants. This swi4-dependent reduction was also observed in mutants additionally defective in msh2 and/or msh6 as well as in swi4 exo1 double mutants. Determination of the mutational spectra of reversions in a swi4 background revealed that the main effects were decreased rates of 2-bp insertions and increased rates of 2-bp deletions.

Msh6 is as important in mismatch repair as Msh2:
Our recent study revealed that swi4 was not significantly different from wild type in repair of base-base mismatches and of one-nucleotide loops, while msh6 showed essentially the same increased mutation rates as msh2 mutants (TORNIER et al. 2001 , accompanying article). Although we found some minor effects on GT repeat stability, the role of Swi4 in repair of small loops is clearly different from that of Msh3 of S. cerevisiae. Another difference is the contribution of Msh6, which is as important as Msh2 in repair of two unpaired nucleotides (this study). In this respect, it would be interesting to study the function of Swi4 and Msh6 in repair of larger loops. However, the results of the GT repeat assays already indicate that repair of four unpaired nucleotides is reduced in msh6 mutants to an extent similar to that in msh2 mutants, but is only marginally affected in swi4 mutants. Consistent with our genetic data is a previous study that showed that purified human MutS is able to bind to loops with up to eight unpaired nucleotides (GENSCHEL et al. 1998 ). In addition, MutS is much more abundant than MutSß (DRUMMOND et al. 1997 ). These findings suggest that MutS is the major MMR-dependent recognition complex for mismatches and loops in human cells. We propose that MutSß in S. pombe and in human rather serves as a backup system and that it might have its main function in other cellular processes like recombination.

Microsatellites in S. pombe:
A number of genetic human disorders are caused by or associated with microsatellite instability (MCMURRAY 1995 ; DEBRAUWERE et al. 1997 ; SIDRANSKY 1997 ). Repeat length variation in genes results in loss of function and in many cases results in diseases. However, the human genome mainly consists of noncoding DNA with a frequent occurrence of microsatellites. Consequently, changes in such repetitive regions do not result in loss of functions. In contrast, the genome of S. pombe is quite compact with a high density of genes. Microsatellites in genes represent high risk factors for losing cellular functions by repeat length variations and are even more critical in haploid organisms like S. pombe. We searched the S. pombe genome released by the S. pombe Sequencing Project (http://www.sanger.ac.uk/Projects/S_pombe/) for the presence of repetitive DNA with units of one or two nucleotides when 83% of the genome sequence was available. We found six mononucleotide runs with unit numbers between 12 and 39 and an enormous number of smaller mononucleotide repeats. Among the larger dinucleotide repeats (>10 units), four GT repeats (the largest consists of 18 units), two AG repeats, and three AT repeats are present in the known part of the genome. Neither one of these microsatellites (the smaller mononucleotides were not analyzed) nor two other repetitive regions containing arrays of dinucleotide repeats, disrupted by other nucleotides, are situated in an open reading frame. Thus, none of the identified microsatellites in S. pombe seems to be a critical factor for loss of gene functions. During evolution such sequences were likely eliminated by mutational changes in S. pombe, while they can be accumulated in a diploid and multicellular organism like us.

	FOOTNOTES

¹ Present address: Department of Genetics, Faculty of Agriculture, Ain Shams University, Shobra El-Khima, P.O. Box 68, 11241 Cairo, Egypt.

	ACKNOWLEDGMENTS

We thank Jürg Kohli who gave A.A.M. the opportunity to work in his lab. This work was supported by the Swiss National Science Foundation. A.A.M. was additionally financed by the Eidgenössische Stipendienkommission für Ausländische Studierende.

Manuscript received May 22, 2000; Accepted for publication January 19, 2001.

	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].

BUERMEYER, A. B., S. M. DESCHÊNES, S. M. BAKER, and R. M. LISKAY, 1999  Mammalian DNA mismatch repair. Annu. Rev. Genet. 33:533-564[Medline].

DEBRAUWERE, H., C. G. GENDREL, S. LECHAT, and M. DUTREIX, 1997  Differences and similarities between various tandem repeat sequences: minisatellites and microsatellites. Biochimie 79:577-586[Medline].

DORIT, R. L., O. OHARA and B.-C. HWANG, 1995 Direct sequencing of PCR products, pp. 15.2.1–15.2.3 in Current Protocols in Molecular Biology, edited by F. M. AUSUBEL, R. BRENT, R. E. KINGSTON, D. D. MOORE, J. G. SEIDMAN, J. A. SMITH and K. STRUHL. John Wiley & Sons, New York.

DRUMMOND, J. T., J. GENSCHEL, E. WOLF, and P. MODRICH, 1997  DHFR/MSH3 amplification in methotrexate-resistant cells alters the hMutS/hMutSß ratio and reduces the efficiency of base-base mismatch repair. Proc. Natl. Acad. Sci. USA 94:10144-10149[Abstract/Free Full Text].

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

FLECK, O., H. MICHAEL, and L. HEIM, 1992  The swi4⁺ gene of Schizosaccharomyces pombe encodes a homologue of mismatch repair enzymes. Nucleic Acids Res. 22:5289-5295[Abstract/Free Full Text].

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].

GENSCHEL, J., S. J. LITTMAN, J. T. DRUMMOND, and P. MODRICH, 1998  Isolation of MutSbeta from human cells and comparison of the mismatch repair specificities of MutSbeta and MutSalpha. J. Biol. Chem. 273:19895-19901[Abstract/Free Full Text].

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].

GRIMM, C., J. KOHLI, J. MURRAY, and K. MAUNDRELL, 1988  Genetic engineering of Schizosaccharomyces pombe. A system for gene disruption and replacement using the ura4 gene as selectable marker. Mol. Gen. Genet. 215:81-86[Medline].

GRIMM, C., J. BÄHLER, and J. KOHLI, 1994  M26 recombinational hotspot and physical conversion tract analysis in the ade6 gene of Schizosaccharomyces pombe.. Genetics 135:41-51.

GUTZ, H., H. HESLOT, U. LEUPOLD and N. LOPRIENO, 1974 Schizosaccharomyces pombe, pp. 395–446 in Handbook of Genetics, Vol. 1, edited by R. C. KING. Plenum Press, New York.

HABRAKEN, Y., P. SUNG, L. PRAKASH, and S. PRAKASH, 1996  Binding of insertion/deletion DNA mismatches by the heterodimer of yeast mismatch repair proteins MSH2 and MSH3. Curr. Biol. 6:1185-1187[Medline].

IACCARINO, I., F. PALOMBO, J. DRUMMOND, N. F. TOTTY, and J. J. HSUAN et al., 1996  MSH6, a Saccharomyces cerevisiae protein that binds to mismatches as a heterodimer with MSH2. Curr. Biol. 6:484-486[Medline].

INOKUCHI, K., M. IKEJIMA, A. WATANABE, E. NAKAJIMA, and H. ORIMO et al., 1995  Loss of expression of the human MSH3 gene in hematological malignancies. Biochem. Biophys. Res. Commun. 214:171-179[Medline].

JIRICNY, J., 1998  Replication errors: cha(lle)nging the genome. EMBO J. 17:6427-6436[Medline].

JIRICNY, J., 2000  Mediating mismatch repair. Nat. Genet. 24:6-8[Medline].

JOHNSON, R. E., G. K. KOVVALI, L. PRAKASH, and S. PRAKASH, 1996  Requirement of the yeast MSH3 and MSH6 genes for MSH2-dependent genomic stability. J. Biol. Chem. 271:7285-7288[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 number of mutants in bacterial populations. J. Genet. 49:264-285.

LIPKIN, S. M., V. WANG, R. JACOBY, S. BANERJEE-BASU, and A. D. BAXEVANIS et al., 2000  MLH3: a DNA mismatch repair gene associated with mammalian microsatellite instability. Nat. Genet. 24:27-35[Medline].

MARRA, G. and P. SCHÄR, 1999  Recognition of DNA alterations by the mismatch repair system. Biochem. J. 338:1-13.

MARRA, G., I. IACCARINO, T. LETTIERI, G. ROSCILLI, and P. DELMASTRO et al., 1998  Mismatch repair deficiency associated with overexpression of the MSH3 gene. Proc. Natl. Acad. Sci. USA 95:8568-8573[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].

MCMURRAY, C. T., 1995  Mechanisms of DNA expansion. Chromosoma 104:2-13[Medline].

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

NAKAGAWA, T., A. DATTA, and R. D. KOLODNER, 1999  Multiple functions of MutS- and MutL-related heterocomplexes. Proc. Natl. Acad. Sci. USA 96:14186-14188[Free Full Text].

PAPADOPOULOS, N., N. C. NICOLAIDES, B. LIU, R. PARSONS, and C. LENGAUER et al., 1995  Mutations of GTBP in genetically unstable cells. Science 268:1915-1917[Abstract/Free Full Text].

RUDOLPH, C., O. FLECK, and J. KOHLI, 1998  Schizosaccharomyces pombe exo1 is involved in the same mismatch repair pathway as msh2 and pms1.. Curr. Genet. 34:343-350[Medline].

RUDOLPH, C., C. KUNZ, S. PARISI, E. LEHMANN, and E. HARTSUIKER et al., 1999  The msh2 gene of Schizosaccharomyces pombe is involved in mismatch repair, mating-type switching, and meiotic chromosome organization. Mol. Cell. Biol. 19:241-250[Abstract/Free Full Text].

SCHÄR, P., M. A. BAUR, C. SCHNEIDER, and J. KOHLI, 1997  Mismatch repair in Schizosaccharomyces pombe requires the mutL homologous gene pms1: molecular cloning and functional analysis. Genetics 146:1275-1286[Abstract].

SCHMUTTE, C., R. C. MARINESCU, M. M. SADOFF, S. GUERRETTE, and J. OVERHAUSER et al., 1998  Human exonuclease I interacts with the mismatch repair protein hMSH2. Cancer Res. 58:4537-4542[Abstract/Free Full Text].

SCHUCHERT, P., M. LANGSFORD, E. KÄSLIN, and J. KOHLI, 1991  A specific DNA sequence is required for high frequency of recombination in the ade6 gene of fission yeast. EMBO J. 10:2157-2163[Medline].

SIA, E. A., R. J. KOKOSKA, M. DOMINSKA, P. GREENWELL, and T. D. PETES, 1997  Microsatellite instability in yeast: dependence on repeat unit size and DNA mismatch repair genes. Mol. Cell. Biol. 17:2851-2858[Abstract].

SIDRANSKY, D., 1997  Nucleic acid-based methods for the detection of cancer. Science 278:1054-1058[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].

SZANKASI, P. and G. SMITH, 1995  A role for Exonuclease I from S. pombe in mutation avoidance and mismatch correction. Science 267:1166-1168[Abstract/Free Full Text].

SZANKASI, P., W. D. HEYER, P. SCHUCHERT, and J. KOHLI, 1988  DNA sequence analysis of the ade6 gene of Schizosaccharomyces pombe. Wild-type and mutant alleles including the recombination hot spot allele ade6-M26. J. Mol. Biol. 204:917-925[Medline].

TISHKOFF, D. X., A. L. BOERGER, P. BERTRAND, N. FILOSI, and G. M. GAIDA et al., 1997  Identification and characterization of Saccharomyces cerevisiae EXO1, a gene encoding an exonuclease that interacts with MSH2. Proc. Natl. Acad. Sci. USA 94:7487-7492[Abstract/Free Full Text].

TISHKOFF, D. X., N. S. AMIN, C. S. VIARS, K. C. ARDEN, and R. D. KOLODNER, 1998  Identification of a human gene encoding a homologue of Saccharomyces cerevisiae EXO1, an exonuclease implicated in mismatch repair and recombination. Cancer Res. 58:5027-5031[Abstract/Free Full Text].

TORNIER, C., S. BESSONE, I. VARLET, C. RUDOLPH, and M. DARMON et al., 2001  Requirement for Msh6, but not for Swi4 (Msh3) in Msh2-dependent repair of base-base mismatches and mononucleotide loops in Schizosaccharomyces pombe.. Genetics 158:65-75[Abstract/Free Full Text].

UMAR, A. and T. A. KUNKEL, 1996  DNA-replication fidelity, mismatch repair and genome instability in cancer cells. Eur. J. Biochem. 238:297-307[Medline].

WANG, T.-F., N. KLECKNER, and N. HUNTER, 1999  Functional specificity of MutL homologs in yeast: evidence for three Mlh1-based heterocomplexes with distinct roles during meiosis in recombination and mismatch correction. Proc. Natl. Acad. Sci. USA 96:13914-13919[Abstract/Free Full Text].

WIERDL, M., M. DOMINSKA, and T. D. PETES, 1997  Microsatellite instability in yeast: dependence on the length of the microsatellite. Genetics 146:769-779[Abstract].

WILSON, D. M., III, J. P. CARNEY, M. A. COLEMAN, A. W. ADAMSON, and M. CHRISTENSEN et al., 1998  Hex1: a new human Rad2 nuclease family member with homology to yeast exonuclease 1. Nucleic Acids Res. 26:3762-3768[Abstract/Free Full Text].

This article has been cited by other articles:

				G.-F. Richard, A. Kerrest, and B. Dujon Comparative Genomics and Molecular Dynamics of DNA Repeats in Eukaryotes Microbiol. Mol. Biol. Rev., December 1, 2008; 72(4): 686 - 727. [Abstract] [Full Text] [PDF]

				T. Yamada-Inagawa, A. J. S. Klar, and J. Z. Dalgaard Schizosaccharomyces pombe Switches Mating Type by the Synthesis-Dependent Strand-Annealing Mechanism Genetics, September 1, 2007; 177(1): 255 - 265. [Abstract] [Full Text] [PDF]

				K. M. Doherty, S. Sharma, L. A. Uzdilla, T. M. Wilson, S. Cui, A. Vindigni, and R. M. Brosh Jr. RECQ1 Helicase Interacts with Human Mismatch Repair Factors That Regulate Genetic Recombination J. Biol. Chem., July 29, 2005; 280(30): 28085 - 28094. [Abstract] [Full Text] [PDF]

				C. Holmberg, O. Fleck, H. A. Hansen, C. Liu, R. Slaaby, A. M. Carr, and O. Nielsen Ddb1 controls genome stability and meiosis in fission yeast Genes & Dev., April 1, 2005; 19(7): 853 - 862. [Abstract] [Full Text] [PDF]

				S. Estes, P. C. Phillips, D. R. Denver, W. K. Thomas, and M. Lynch Mutation Accumulation in Populations of Varying Size: The Distribution of Mutational Effects for Fitness Correlates in Caenorhabditis elegans Genetics, March 1, 2004; 166(3): 1269 - 1279. [Abstract] [Full Text] [PDF]

				C. Tornier, S. Bessone, I. Varlet, C. Rudolph, M. Darmon, and O. Fleck Requirement for Msh6, but Not for Swi4 (Msh3), in Msh2-Dependent Repair of Base-Base Mismatches and Mononucleotide Loops in Schizosaccharomyces pombe Genetics, May 1, 2001; 158(1): 65 - 75. [Abstract] [Full Text]

				M. Hohl, O. Christensen, C. Kunz, H. Naegeli, and O. Fleck Binding and Repair of Mismatched DNA Mediated by Rhp14, the Fission Yeast Homologue of Human XPA J. Biol. Chem., August 10, 2001; 276(33): 30766 - 30772. [Abstract] [Full Text] [PDF]