The Yeast HSM3 Gene Acts in One of the Mismatch Repair Pathways

Corresponding author: Vladimir G. Korolev, Petersburg Nuclear Physics Institute, Division of Molecular and Radiation Biophysics, Russian Academy of Science, 188350, Gatchina, Leningrad District, Russia, lge{at}omrb.pnpi.spb.ru (E-mail).

Communicating editor: G. R. SMITH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutants with enhanced spontaneous mutability (hsm) to canavanine resistance were induced by N-methyl-N-nitrosourea in Saccharomyces cerevisiae. One bearing the hsm3-1 mutation was used for this study. This mutation does not increase sensitivity to the lethal action of different mutagens. The hsm3-1 mutation produces a mutator phenotype, enhancing the rates of spontaneous mutation to canavanine resistance and reversions of lys1-1 and his1-7. This mutation increases the rate of intragenic mitotic recombination at the ADE2 gene. The ability of the hsm3 mutant to correct DNA heteroduplex is reduced in comparison with the wild-type strain. All these phenotypes are similar to ones caused by pms1, mlhl, and msh2 mutations. In contrast to these mutations, hsm3-1 increases the frequency of ade mutations induced by 6-HAP and UV light. Epistasis analysis of double mutants shows that the PMS1 and HSM3 genes control different mismatch repair systems. The HSM3 gene maps to the right arm of chromosome II, 25 cM distal to the HIS7 gene. Strains that bear a deleted open reading frame YBR272c have the genetic properties of the hsm3 mutant. The HSM3 product shows weak similarity to predicted products of the yeast MSH genes (homologs of the Escherichia coli mutS gene). The HSM3 gene may be a member of the yeast MutS homolog family, but its function in DNA metabolism differs from the functions of other yeast MutS homologs.

CELLS have evolved a number of mechanisms to ensure high-fidelity transmission of genetic material from one generation to the next, since mutations can lead to genotypes that may be deleterious to the cell. There are at least three ways in which mismatched nucleotides arise in DNA. First, physical damage to the DNA can give rise to mismatched bases (FRIEDBERG 1985 ). Second, misincorporation of nucleotides during DNA replication can yield mismatched basepairs, nucleotide insertions, and deletions (MODRICH 1991 ). Finally, genetic recombination produces regions of heteroduplex DNA that may contain mismatched nucleotides arising from the pairing of two different parental DNA sequences (HOLLIDAY 1964 ). Mismatched nucleotides produced by each of these mechanisms are repaired by specific enzyme systems (FRIEDBERG 1990 ; MODRICH 1991 ).

DNA mismatch repair is responsible for the correction of replication errors and deaminated cytosines. The lack of mismatch repair usually increases mutation rates and causes a mutator phenotype (COX 1976 ).

In both prokaryotes and eukaryotes, DNA mismatch repair plays a prominent role in the correction of errors made during DNA replication and genetic recombination (GRILLEY et al. 1990 ; RADMAN 1988 ). In Escherichia coli, methyl-directed mismatch repair involves the products of the mutator genes mutS, mutL, mutH, and uvrD (MODRICH 1991 ). Isolation of the respective proteins has allowed characterization of their biochemical functions: MutS is a DNA mismatch-binding protein (SU et al. 1988 ; SU and MODRICH 1986 ), UvrD is DNA helicase II (HICKSON et al. 1983 ), and MutH is a latent endonuclease that incises at the transiently unmethylated strands of hemimethylated GATC sequences (WELSH et al. 1987 ). MutL acts as a "molecular matchmaker," a protein that promotes the formation of a stable complex between two or more DNA-binding proteins in an ATP-dependent manner (SANCAR and HEARST 1993 ).

Elements of the E. coli MutLHS system appear to have been evolutionarily conserved in prokaryotes and eukaryotes. The first molecular evidence for the conservation of DNA mismatch repair in yeast came from the characterization of the Saccharomyces cerevisiae PMS1 gene, the protein product of which exhibits similarity to MutL (KRAMER et al. 1989A ; WILLIAMSON et al. 1985 ). Unlike bacterial systems, however, the S. cerevisiae pathway has two MutL homologs, PMS1 (KRAMER et al. 1989A ) and MLH1 (PROLLA et al. 1994 ), that act in the same pathway. Six S. cerevisiae DNA mismatch repair genes MSH1, MSH2, MSH3, MSH4, MSH5, and MSH6 have been identified and found to encode proteins displaying homology to prokaryotic MutS (REENAN and KOLODNER 1992A ; NEW et al. 1993 ; ROSS-MACDONALD and ROEDER 1994 ; HOLLINGSWORTH 1995 ; DRUMMOND et al. 1995 ; MARSISCHKY et al. 1996 ).

Recent studies have revealed that a number of tumor typed (particularly hereditary nonpolyposis colon cancer) are associated with high-frequency alterations of microsatellite sequences (AALTONEN et al. 1993 ; THIBODEAU et al. 1993 ; IONOV et al. 1993 ; LINDBLOM et al. 1993 ; HAN et al. 1993 ). In some families, this instability is associated with mutations in hMSH2, hMLH1, hPMS1, and hPMS2 genes that encode human homologs of the E. coli mismatch-repair proteins MutS and MutL (FISHEL et al. 1993 ; LEACH et al. 1993 ; LI and MODRICH 1995 ; PAPADOPOULOS et al. 1994 ; NICOLAIDES et al. 1994 ).

Earlier, we isolated a number of mutator mutants with high spontaneous mutagenesis (hsm mutants; IVANOV et al. 1992 ). In this paper, we described the hsm3 mutant, which has a strong mutator effect, and studied its properties to determine the role it plays in mutagen-induced mutagenesis and mismatch correction.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains:
S. cerevisiae strains used in this paper are listed in Table 1.

Mutagens:
N-methyl-N-nitrosourea (MNU) was a kind gift from R. I. SALGANIK. 6-N-hydroxylaminopurine (6-HAP) was synthesized in our Institute by G. N. BONDAREV. Methylmethansulfonate (MMS) was received from Sigma (St. Louis, MO). The UV light source was a BUV-30E lamp with a dose rate of 2.76 J/m²/s. Gamma irradiation was from ⁶⁰Co with a dose rate of 69 Gy/min.

Mutation assays:
The fluctuation test of the median (LEA and COULSON 1949 ) was used to determine the spontaneous mutation rate to canavanine resistance in the original strain ELI-1A-a116 bearing the hsm3 mutation and in the wild-type strain ELI-1B-a116. In these experiments, 20 cultures, which were started with 100–150 cells, were used. In other experiments, the spontaneous mutation rate to canavanine resistance, reversions for alleles his1-7 and lys1-1, and the frequency of intragenic recombination in the ADE2 locus were determined by KHROMOV-BORISOV's method of ordered plating (VON BORSTEL 1978 ) as follows. The tested yeast cultures were grown on plates with complete medium for 1 day, and then 5 ml of a suspension (10⁶ cells/ml) was prepared. A special 150-stamp replicator was dipped into this suspension and inverted on a plate with selective medium containing a limiting concentration of amino acids (2 mg/liter of histidine and 4 mg/liter lysine in the experiments on the study of spontaneous reversion rates, 2 mg/liter adenine in the experiments on the study of spontaneous conversion rates, or 50 mg/liter canavanine and 5% liquid YEPD medium in experiments on the mutation frequency to canavanine resistance). The replicator places 150 equal drops of yeast suspension (~2 µl each) at equal distances from each other. In the experiments, each drop contained ~2000 cells. The mutants and convertants have faster growth that shows up as papillae on the spots with limited growth of the tested culture. After 14–15 days of incubation, the papillae and number of cells were counted. The latter was done after washing the cells from the entire plate or from a number of single-drop replicas lacking papillae. The yeast grow uniformly when plated in this manner. This minimizes the variation of frequencies (VON BORSTEL 1978 ). If the mutant cells preexisted in the culture before being plated, the papillae arose within 3 days after inoculum plating and had the same size. The experiments were ignored. The rates of mutation per cell division were determined by dividing the number of papillae by the total number of cells on a plate.

The investigation of forward-induced mutations:
Forward adenine-requiring mutants carrying mutations in the ADE1 or ADE2 locus were induced by UV, 6-HAP, MNU in hsm3, and HSM3 strains. In experiments on the study of UV-induced mutagenesis in double mutants, Roman's mutation system was used on five ADE genes (ROMAN 1956 ). For this purpose, the strains carrying ade2 were used. Special media, excluding petite mutant growth, were used.

Mutation induction by UV light, 6-HAP, and MNU:
These were performed as described (IVANOV et al. 1989 ).

Analysis of spontaneous mitotic gene conversion at the ADE2 locus:
A set of diploid strains heteroallelic at the ADE2 gene was constructed. One mating partner had the genotype MATa ade2-58 ura4 hsm3-1 (or HSM3). The other partner was MAT ade2-i trpl hsm3-1, with ade2-i being independent UV-induced mutations obtained in strain 2C-SVK-165. Diploid strains were selected on minimal plates supplemented with adenine (50 mg/liter). Freshly isolated hybrids were plated using a 150-stamp replicator on minimal plates with limiting adenine (2 mg/liter), as proposed by KHROMOV-BORISOV (VON BORSTEL 1978 ). White colonies (papillae) of convertants were scored after 14 days of incubation at 30°, and conversion rates were expressed as the number of white colonies per cell division.

The construction of double-mutant hsm3 pms1:
The method of construction of the PMS1 deletion mutants was described (KRAMER et al. 1989B). Integrative plasmid YIP5 carrying the URA3 gene and a fragment with the pms1 deletion was a kind gift from Dr. W. KRAMER. The plasmid DNA was linearized with HpaI and used for the transformation of strain 21VF-313 to uracil prototrophy. Intrachromosomal Ura^- recombinants were selected on plates containing 5-fluoroorotic acid. Because both the pms1 and hsm3 mutations increased the mutation rates to canavanine resistance (WILLIAMSON et al. 1985 ), isolates bearing both mutations were identified as follows: Sixteen putative double mutants were crossed with a pms1 strain (2IVF-312). The 11 resulting diploids showed an increased mutation rate to canavanine resistance; i.e., they were homozygous for pms1 and, consequently, the original haploids carried the hsm3 and pms1 mutations.

The construction of plasmids:
To make vectors for hetero-duplex transformation experiments, we used shuttle centromer-ic vectors pFL59⁺ and pFL59^- with the opposite orientation of the ori f1 region (BONNEAUD et al. 1991 ). The PstI- EcoRI fragment of plasmid pMB20 (CHEPURNAYA et al. 1993 ) containing the ADE2 gene was inserted in the polylinker of pFL59⁺ and pFL59^- to form our pFLA590⁺ and pFLA590^- plasmids (Figure 1). These plasmids allowed us to produce separate, single-stranded DNAs that carried either the coding (pFLA591⁺) or antisense (pFLA591^-) strand of ADE2 and that were almost complementary (with the exception of the small ori f1 region, which was distant form ADE2).

View larger version (16K): In this window In a new window Download PPT slide   Figure 1. —Structure of double-stranded plasmids pFLA590+ and pFLA590-. The arrows indicate the direction of replication from the F1 origin that leads to appearance of + or - single-stranded plasmids.

To produce plasmids with a point mutation (GC CG) in ADE2, we made a 1249-bp DNA fragment of the ADE2 gene with a one-nucleotide substitution in the starting methionine codon using PCR with the following primers: HTG1 primer without substitution (5' atgctgatcccatgatgatt-3') and HTG2 primer with G to C substitution (underlined) in the ATG codon (5'-acaaaacaatcaagtatcgatt-3') and pFLA590⁺ as a template. This fragment was annealed to single-stranded pFLA591⁺ and pFLA591^- plasmids with slow cooling from 94° to 50° in the presence of 0.3 M NaCl, which was added after 10 min of heating at 94°. Such plasmids with the annealed fragments were used for yeast transformation (see below). Plasmid DNAs from red colonies of yeast transformants were transferred to E. coli by the fast method (MARCIL and HIGGINS 1991 ). Ampicillin-resistant colonies of E. coli transformants were taken to produce large amounts of single-stranded plasmids with the point mutations in ADE2 (pFLA592⁺ carried G C substitution and pFLA592^- carried C G substitution). All the nucleotide substitutions were verified by sequencing.

The construction of artificial heteroduplex:
We annealed two complementary single-stranded plasmids, one of which had a one-nucleotide substitution in the starting codon of the ADE2 gene, so that a mismatched nucleotide pair was formed. One of these single-stranded plasmids had been previously restricted by XbaI (at two sites) and PvuII (at one site) to three fragments using the following oligonucleotides to form double-stranded DNA regions in restriction sites: HDR1, 5'-tggaaaagatgctagaccta-3'; HDR2,5'-taggtctagactcttttca-3'; HDR3, 5'-gcatcggaatctagagcaca-3'; HDR4, 5'-tgtgctctagattccgatgc-3'; HDR5, 5'-gctattacgccagctggcga-3'; HDR6, 5'-tcgccagctggcgtaatagc-3'. HDR1 and HDR2 formed an XbaI site in ADE2 (1558 position), and HDR1 in the coding and HDR2 in antisense strands, respectively. HDR3 and HDR4 formed an XBAI site in the vector, HDR3 in - plasmids and HDR4 in + plasmids; HDR5 and HDR6 formed a PvuII site in the vector, HDR5 in - and HDR6 in + plasmids. A 10-fold molar excess of three of these oligo-nucleotides was added to single-stranded plasmid DNA; and was annealed by heating 60° for 10 min and fast cooling in the presence of 0.01 M MgCl₂ and 0.05 M NaCl. The plasmid with annealed oligonucleotides was precipitated by ethanol in the presence of 0.3 M sodium acetate to remove unused oligonucleotides, was dissolved in 10 mM Tris/1 mM Na₂ EDTA (pH 8), and was restricted in standard restriction buffers.

The restricted strand was annealed to the nonrestricted strand during 4 hr with slow cooling from 94° to 50° in the presence of 0.3 M NaCl, which was added after 10 min heating at 94°. The annealing mixture (100 µl) contained 2 µg of nonrestricted strand DNA and 8 µg of restricted strand DNA. After annealing, this DNA was precipitated with ethanol and dissolved in 10 µl of 10 mM Tris/1 mM Na₂ EDTA (pH 8).

The construction of deleted open reading frame YBR272c:
Two 21-mer deoxyoligonucleotides, ETA1 (5'-TTCTTCACCTTTGTCCAGTAA-3') and ETA2 (5'-CGTCTAACGTATCCTATGATT-3'), were synthesized. S. cerevisiae genomic DNA was provided by S. A. BULAT. A PCR was carried out in a 20-µl volume containing 20 mM Tris-HCl (pH 8.8), 10 mM (NH₄)₂SO₄, 0.6 mM each of dNTPs, 4 mM MgCl₂, 40 ng of single primers, between 20 and 200 ng of the yeast genomic DNA, and 2–3 units of Tsp DNA polymerase (obtained from O. K. KABOEV). PCR was performed using the thermal cycler TC-1000M (PNPI, St. Petersburg, Russia) for 30 cycles. The cycling parameters were as follows: DNA denaturation at 92° for 50 sec (first denaturation step at 94° for 2.5 min), primer extension at 69° for 60 sec, and primer annealing at 52–55° for 90 sec. During the first five cycles, the primer annealing time was lengthened to 120 sec. The final extension step took place at 69° for 3 min. Amplification products were analyzed in 1.7% agarose gels at 150 V with cooled TBE buffer for 40 min and then visualized by staining with ethidium bromide. A fragment of the expected size (~1000 bp) was obtained. To create a deletion of the open reading frame (ORF), YBR272c, the 1-kb PCR product was cloned into plasmid pFL34 (3.88 kb). This deleted construction was linearized by the Dral I enzyme and transformed into the 11D-3021 strain. Ura⁺ transformants were tested for spontaneous and UV-induced mutagenesis.

Genetic procedures:
Standard methods of hybrid selection, sporulation, micromanipulation, and tetrad analysis were as described in ZAKHAROV et al. 1984 . Yeast transformations were performed by the lithium acetate method (ITO et al. 1983 ). Standard molecular biology techniques described by MANIATIS et al. 1982 were used.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To isolate mutants with enhanced spontaneous mutability, the strain 197/2 was treated with MNU. A collection of strains with the desired phenotype was isolated (IVANOV et al. 1992 ). They were named hsm (high spontaneous mutagenesis). Five of these strains were crossed with strain 754, and tetrad analysis of the hybrids showed that the character of enhanced mutability segregated 2:2. Recombinational tests demonstrated that the five tested hsm strains carried five nonallelic mutations (IVANOV et al. 1992 ). One of these mutants, hsm3, was used for the following study.

Influence of the hsm3-1 mutation on the sensitivity to the lethal action of mutagens:
The study of the sensitivity to the lethal effects of UV light, 6-HAP, MNU (Figure 2), MMS, and rays (data not presented) showed that the hsm3 mutant did not differ from the wild-type strain. The hsm3-1 mutation thus does not affect the repair of lethal DNA damage.

View larger version (62K): In this window In a new window Download PPT slide   Figure 2. —Induction of ade1 and ade2 mutants and the viability of wild-type and hsm3 strains after exposure to various mutagens. Curves of viability and mutant frequency represnt an average of four different experiments for each strain: (A) by UV light, (B) by 6-HAP (6-hr exposure); (C) by MNU (1.5 hr exposure). Survival: hsm3, +; wild type, *. Mutagenesis: hsm3, x; wild type, .

Effects of the hsm3 on spontaneous mutagenesis:
Forward mutations to canavanine resistance and reversions of lys1-1 and his1-7 were used to estimate spontaneous mutability (Table 2). The forward mutation rates to canavanine resistance were determined in the original mutant strain ELI1A-a116; strain ELI1B-a116 was used as control. The data shown in this table indicates that the hsm3 strain displays a spontaneous mutator phenotype (increasing the spontaneous mutation rate by ~10-fold). The rates of spontaneous reversions of lys1-1 (nonsense mutation) and his1-7 (missense mutation) in the haploid hsm3 mutant (IVF-1C) and in diploids homo- and heterozygous for hsm3 and homozygous for lys1-1 and his1-7 (IVF-D2 and IVF-D3) were determined. Mutant hsm3-1 increases the reversion rates by ~7–10-fold in the haploids and the diploids. The mutation reversion rates of diploids heterozygous for hsm3-1 did not differ from that of wild-type diploids. These data show that the hsm3-1 mutation is recessive.

Effects of the hsm3-1 mutation on spontaneous mitotic gene conversion at the ADE2 locus:
To learn more about the participation of the HSM3 gene in various pathways of DNA metabolism, we studied the effects of the hsm3-1 mutation on spontaneous mitotic gene conversion at the ADE2 locus. To study intragenic recombination, special strains were constructed. Twelve ade2 mutations were induced by UV light in the hsm3 mutant (2C-SVK-165). These ade2-i mutants were crossed to 5SVK-166 (hsm3-1 ade2-58) and p3058 (HSM ade2-58) strains. Allele ade2-58 is an ochre nonsense mutation located near the 3' end of ADE2 (KOVALTZOVA and KOROLEV 1989 ). Diploid strains that were heterozygous (HSM3/hsm3) or homozygous (hsm3/hsm3) and maintained ade2-58/ade2-i heteroallelic combinations were constructed. Data on eight heteroallelic combinations are presented in Table 3. In most cases, the hsm3-1 mutation increased the frequency of mitotic intragenic conversion by approximately twofold.

Effects of the hsm3-1 on induced mutagenesis:
We used 6-HAP, MNU, and UV light for induction of ade1 and ade2 mutations in the hsm3 mutant (ELI1A-a116). Strain ELI1B-a116 was used as a control. The hsm3-1 mutation increased cell sensitivity to mutagenic UV light action (Figure 2A).

6-HAP, a highly efficient chemical mutagen, is an analog of purines. The effect of hsm3-1 on forward mutations in the ADE1 and ADE2 genes induced by 6-HAP was determined (Figure 2B). The hsm3 mutant showed greater sensitivity to the mutagenic effect of 6-HAP than does the wild-type strain. When forward MNU-induced mutations in the ADE1 and ADE2 genes were examined, however, no significant differences between the strains under investigation were observed (Figure 2C). It is important to note that the hsm3 strain showed no increased sensitivity to the lethal effects of UV light, 6-HAP, and MNU.

Effects of hsm3 mutation on the spontaneous mutability of diploids:
The major mechanism behind the appearance of UV- and propiolactone-induced diploid mutants in S. cerevisiae is a mutation in one of the two wild-type alleles and subsequent mitotic homozygotization of the mutations as a result of mitotic segregation including mitotic gene conversion (GORDENIN and INGE-VECHTOMOV 1981 ; PAVLOV et al. 1988 ). Based on these data, we assumed that in diploids homozygous for mutator mutations, the spontaneous mutation frequency to canavanine resistance (a recessive mutation) would be considerably higher than in diploids heterozygous for mutator mutations or in the wild type because the spontaneous mutator mutation would enhance both the probability of a can^R mutation appearing in one of the chromosomes and the probability of homozygotization for can^R.

Experimental data confirmed our hypothesis. We incubated the diploids homozygous or heterozygous for hsm3 and pms1, as well as wild-type diploids in liquid YEPD medium for 48 h at 30°. Fifty microliters of the suspensions (~10⁸ cells/ml) was then plated on selective media with 50 mg/liter canavanine and incubated for 10 days. The number of canavanine-resistant clones in hsm3/hsm3 and pms1/pms1 diploids was ~25–30, but none appeared in heterozygous or wild-type diploids.

We used this assay to show evidence of hsm3-1 and hsm3-1 homozygosis, as well as in the experiments on the construction of the double mutants hsm3 pms1.

Genetical mapping of the HSM3 gene:
By the method of chromosome destabilization (FALCO and BOTSTEIN 1983 ), it was found that the HSM3 gene is located on the right arm of chromosome II (data not shown). From Table 4, one can see that the HSM3 gene is located near HIS7, and the map distance of HSM3 from HIS7 is ~25 cM according to the calculation of PERKINS 1949 , or 25.1 cM if the linkage was calculated taking chromatid interference into consideration (MA and MORTIMER 1983 ).

Physical mapping of the HSM3 gene:
Analysis of DNA sequences in the GenBank database revealed an ORF, YBR272c, encoding a protein that showed weak homology to the MSH1 gene product (FELDMAN et al. 1994 ). This gene was located 32 kb distally from HIS7. According to genetic mapping, HSM3 was located 25 cM distally from HIS7. Epistasis analysis and the measurement of spontaneous mutation rate and heteroduplex repair suggest that HSM3 may act in a novel mismatch repair pathway responsible for the removal of mispaired bases in DNA. We supposed, therefore, that HSM3 is ORF YBR272c.

We deleted ORF YBR272c in strain 11D-3031 and compared the resulting mutant with the hsm3 mutant. The spontaneous mutation rate to canavanine resistance and the frequency of UV-induced mutations in five ADE genes were determined for the deletion mutant (Table 5). The data of this table show that the phenotypical properties of the deletion mutant and the original hsm3 mutant are identical. Strain 3IVF-314, which carries the hsm3-1 mutation, was crossed with the 2LMG-316 strain which bears the deletion mutation, and the frequency of spontaneous mutations to canavanine resistance was determined in this diploid. This diploid was characterized by increased spontaneous mutability (Table 5). The diploid heterozygous for hsm3-1 from crossing ELI1B-a116 with 2LMG-316 strains was characterized by its low level of canavanine resistance mutation rate (Table 5). These data show that hsm3-1 is allelic to hsm3-1 and that hsm3-1 is recessive. The data on allelism of hsm3-1 and hsm3-1 were also obtained from tetrad analysis of the diploid from crossing 3IVF-314 with 2LMG-316. The mutator phenotype segregated 4 Mut^-:0 Mut⁺ in all cases (Table 4), the hsm3 mutation is allelic to the deletion mutation. We mapped the URA3 insertion in the ORF YBR272c relative to HIS7. The results show that the URA3 insertion is mapped at the same distance as hsm3-1 (Table 4).

According to the GenBank database, the Hsm3 amino acid sequence shows weak homology to the S. cerevisiae MSH1 gene product (FELDMAN et al. 1994 ). The properties of the hsm3 mutant suggest that HSM3 controls one of the mismatch repair systems that takes part in the repair of spontaneous and induced premutation DNA damage. The evolutionary relationship among the known MutS-related proteins indicates that HSM3 belongs to a group of MutS homologs.

Correction of DNA heteroduplex:
The results described above suggest that HSM3 can take part in mismatch repair. We studied therefore the repair of artificial heteroduplex DNA carrying C:C and G:G mispairs (see MATERIALS AND METHODS) in 2-LMG-316 (hsm3-1), 2IVF-312 (pms1), and 1LMG-304 (wild-type) strains. The last two were used as positive and negative controls. The frequency of heteroduplex repair was estimated by the fraction of mosaic colonies among all transformants; this fraction increases with a decrease of heteroduplex correction efficiency. In all strains, C:C heteroduplex repair is less frequent than repair of G:G (Table 6). In the pms1 and hsm3 mutants, repair of the C:C heteroduplex was five and two times less frequent than in the wild-type strain, respectively. The repair of G:G in these mutants was 10 and three times less frequent than in the wild type, respectively. Differences in the fraction of mosaic colonies among pms1, hsm3, and wild-type strains were statistically significant (Table 6). These data show that the pms1 and hsm3 mutations reduce the ability of cells to correct artificial heteroduplexes. To study the influence of nicks in one or another strand of heteroduplex DNA on its preference in the repair process, we transformed the cells of pms1 and hsm3 mutants and the wild-type strain with heteroduplex DNA (G:G) carrying nicks in one or another strand. The data show that nicks are not signals for preferential repair of heteroduplex DNA strands in these strains (Table 6).

Epistasis analysis:
The product of the PMS1 gene takes part in mismatch repair and correction of artificial DNA heteroduplex (KRAMER et al. 1989B ). It was therefore very interesting to study the interaction of the PMS1 and HSM3 genes in DNA heteroduplex correction. To determine whether these gene products function in the same pathway of mismatch repair, we constructed a hsm3 pms1 double mutant and measured the spontaneous and UV-induced mutation rates in four independent experiments. The double mutant showed an additive response. The forward spontaneous mutation rates to canavanine resistance were as follows: in hsm3 pms1, (99.1 ± 9) x 10^-7; hsm3, (40.2 ± 4) x 10^-7; and pms1 (60.2 ± 5) x 10^-7. The double mutant was also more sensitive to the mutagenic action of UV light than were the single mutants (Figure 3). These data suggest that HSM3 and PMS1 operate in different mismatch repair pathways.

View larger version (19K): In this window In a new window Download PPT slide   Figure 3. —Induction of mutants in five ADE loci by UV light. Curves of mutant frequency represent an average of three different experiments for each strain: hsm3, ; pms1, *; hsm3 pms1, x; wild type, +.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutations in the HSM3 gene lead to a nuclear mutator phenotype as the spontaneous rates of accumulation of canavanine-resistant mutants and reversion rates of lys1-1 and his1-7 were increased (7–10-fold; Table 2). Tetrad analysis showed that the character of enhanced mutability is monogenic. Homozygous hsm3-1 diploids had the same level of spontaneous reversions as haploid strains, and heterozygous diploids do not differ from the wild-type strain (Table 2). hsm3-1 is therefore a recessive nuclear mutation. The resistance of the hsm3 mutant to the lethal effects of number mutagens was the same in the wild-type strain (Figure 2). These data show that the HSM3 gene does not take part in the repair of lethal DNA damage.

The effect of the hsm3 mutation on mitotic gene conversion events at the ADE2 locus suggest that the HSM3 gene product is involved in the repair of mispaired bases presented in recombination intermediates. All the phenotypes caused by hsm3 mutations are similar to the phenotypes caused by mutations in the PMS1, MLH1, MSH2, and HIM1 genes of yeast (BISHOP et al. 1987 , BISHOP et al. 1989 ; KRAMER et al. 1989A , KRAMER et al. 1989B ; REENAN and KOLODNER 1992A ; IVANOV et al. 1989 ). We assume, therefore, that the HSM3 and HIM1 genes belong to the group that controls mismatch repair in yeast.

We transformed yeast cells with plasmid heteroduplex DNAs, each containing defined mismatches. We found that these mismatches were repaired efficiently in the wild-type cells. Repair of these mismatches, however, severely impaired in the hsm3 mutant and in the pms1 mismatch repair mutant. This result demonstrates that the products of the HSM3 and PMS1 genes are involved in DNA mismatch correction. This is in accordance with the mitotic mutator phenotype and the increased frequency of mitotic gene conversion.

In contrast to the other mismatch repair mutations, hsm3-1 increased not only spontaneous mutations, but also mutations induced by 6-HAP and UV light. 6-HAP is highly mutagenic in growing yeast cultures (PAVLOV et al. 1991 ). Most of the genetic and biochemical evidence suggests that 6-HAP provokes mutations during DNA synthesis because of its ambivalent pairing capacity. dHAP-triphosphate replaces both dATP and dGTP during in vitro DNA synthesis by prokaryotic and eukaryotic DNA polymerases (ABDUL-MASIH and BESSMAN 1986 ). Consistent with this ambiguous pairing, 6-HAP induces both GC AT and AT GC transitions in yeast (SHCHERBAKOVA and PAVLOV 1993 ; NOSKOV et al. 1994 ). The pairs formed in this process, 6-HAP:C and 6-HAP:T, can be recognized by mismatch repair enzymes. The increased 6-HAP–induced mutagenesis observed in the hsm3 mutant may be caused by the inability to correct mismatches appearing in DNA after incorporation of the analog. E. coli mismatch–deficient mutants also show hyper mutability induced by the base analog 2-aminopurine (GLICKMAN and RADMAN 1980 ).

Increased UV-induced mutagenesis in the hsm3 mutant may be caused by the formation of mismatch-like structures that appear during the course of UV light–induced DNA damage repair. The absence of any differences in MNU-induced mutagenesis between the hsm3 mutant and the wild-type strain can be explained by the inability of alkylating mutagens to induce DNA damage of the mismatch type.

Our data suggest that the mismatch repair system controlled by PMS1 differs from that controlled by HSM3. The confirmation of this assumption was received by the study of spontaneous and UV-induced mutagenesis in the double mutant hsm3 pms1. These mutation have an additive type of interaction for spontaneous mutagenesis and a synergistic type for UV-induced mutagenesis, i.e., HSM3 and PMS1 genes are the ones that control the different repair systems (Figure 4).

View larger version (52K): In this window In a new window Download PPT slide   Figure 4. —Sequence homology between Hsm3 and Msh1. The deduced amino acid sequence of Hsm3 was compared with that of Msh1. For the two protein sequences, the straight lines indicate identical amino acids, and dotted lines indicate similar amino acids (double dots are more similar than single dots).

A yeast ORF YBR272c was identified as part of the project to sequence the entire S. cerevisiae chromosome II (FELDMAN et al. 1994 ). We constructed strains bearing deleted ORF YBR272c and showed that all genetic properties of these strains were identical to those of the hsm3 mutant, and the deletion mutation was mapped at the same distance from the HIS7 gene as mutation hsm3-1. So we have identified the yeast gene, HSM3, whose putative gene product shows a weak degree of similarity to the predicted product of the yeast MSH1 gene, which functions in mismatch repair of mitochondrial DNA (REENAN and KOLODNER 1992A , REENAN and KOLODNER 1992B ). Using the FASTA search and the PileUp programs of the Genetics Computer Group (version 8.0; DEVEREUX et al. 1984 ), we found similarity between the Hsm3 protein and the other yeast homologs of MutS. The sequences were Msh1 and Msh2 (REENAN and KOLODNER 1992A ), Msh3 (NEW et al. 1993 ), Msh4 (ROSS-MACDONALD and ROEDER 1994 ), Msh5 (HOLLINGSWORTH et al. 1995), and Msh6 (MARSISCHKY et al. 1996 ). The results showed percentages of similarity and identity with the Msh1 protein of 45.6 and 18.8% (Figure 4); with Msh2, 46.7 and 20.1%; with Msh3, 44.5 and 20.3%; with Msh4, 48.0 and 22.4%, respectively. The predicted molecular mass of the Hsm3 gene product (55.5 kD) is smaller than that of the other yeast Msh (Msh1–Msh6) proteins and procaryotic MutS proteins (95 kD).

MSH1 is involved in yeast mitochondrial DNA repair, as indicated by the high percentage of petites in the msh1 mutant population. MSH2 appears to be a mismatch repair gene for nuclear DNA, as indicated by a highly elevated mutation rate to canavanine resistance and an increase in postmeiotic segregation in msh2 mutant strains (REENAN and KOLODNER 1992A , REENAN and KOLODNER 1992B ). MSH3 does not substantially affect the spontaneous mutation rate at several loci, but it increases microsatellite instability ~40-fold, preferentially causing deletions (NEW et al. 1993 ; STRAND et al. 1995 ). Deletion of the MSH4 gene has no apparent effect on mismatch repair. Msh4 mutant strains display wild-type levels of gene conversion and postmeiotic segregation, but they show a reduction in crossing over and a resultant increase in nondisjunction of homologous chromosomes at meiosis I (ROSS-MACDONALD and ROEDER 1994 ). The product of MSH5 is also a meiosis-specific protein with a role in resolving recombination intermediates rather that a role in mismatch correction of heteroduplex recombination intermediates (HOLLINGSWORTH 1995 ). MSH6 is a homolog of the human gene GTBP that, when defective, leads to genetically unstable cells (DRUMMOND et al. 1995 ).

It is possible, therefore, that the HSM3 gene is a member of yeast MutS homolog family, but its function in DNA metabolism differs from the functions of other yeast MutS homologs. The existence of seven yeast MutS homologs confirms the existence of multiple mismatch repair pathways in yeast (MARSISCHKY et al. 1996 ).

	ACKNOWLEDGMENTS

We thank S. BULAT for preparing the PCR fragments, S. KOZHIN for help in mapping, and V. BASHKIROV for the mathematical treatment of data for MSH homologs. We thank J. KOHLI for critical reading of the manuscript. This work was supported by the grants from The Russian Fund of Fundamental Research and Russian State Program Frontiers in Genetics.

Manuscript received July 17, 1996; Accepted for publication November 24, 1997.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AALTONEN, L. A., P. PELTOMAKI, F. S. LEACH, P. SISTONEN, and L. PYLKKANEN et al., 1993  Clues to the pathogenesis of familial colorectal cancer. Science 260:812-816[Abstract/Free Full Text].

ABDUL-MASIH, M. T. and M. J. BESSMAN, 1986  Biochemical studies on the mutagen 6-N-hydroxylaminopurine. Synthesis of the deoxynucleotide triphosphate and its incorporation into DNA in vitro. J. Biol. Chem. 261:2020-2026[Abstract/Free Full Text].

BISHOP, D. K., M. S. WILLIAMSON, S. FOGEL, and R. D. KOLODNER, 1987  The role of heteroduplex correction in gene conversion in Saccharomyces cerevisiae. Nature 328:362-364[Medline].

BISHOP, D. K., J. ANDERSEN, and R. D. KOLODNER, 1989  Specificity of mismatch repair following transformation of Saccharomyces Cerevisiae with heteroduplex plasmid DNA. Proc. Natl. Acad. Sci. USA 86:3713-3717[Abstract/Free Full Text].

BONNEAUD, N., O. ORIER-KOLOGEROPULOS, G. LI, M. LABOUESSE, and L. MINVIELLE-SEBASTIA et al., 1991  A family of low and high copy replicative, integrative and single-stranded S. cerevisiae/E. coli shuttle vectors. Yeast 7:609-615[Medline].

CHEPURNAYA, O. V., T. N. KOZHINA, V. T. PESHEKHONOV, and V. G. KOROLEV, 1993  REC41—a new gene involved in the control of recombination in the yeast Saccharomyces cerevisiae. Genetika (Moscow) 29:246-256[Medline].

COX, E. C., 1976  Bacterial mutator genes and the control of spontaneous mutation. Annu. Rev. Genet. 10:135-156[Medline].

DEVEREUX, J., P. HAEBERLI, and O. SMITHIES, 1984  A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12:387-395.

DRUMMOND, J. T., G.-M. LI, M. J. LONGLEY, and P. MODRICH, 1995  Isolation of an hMSH2-p160 heterodimer that restores DNA mismatch repair to tumor cells. Science 268:1909-1912[Abstract/Free Full Text].

FALCO, S. A. and D. A. BOTSTEIN, 1983  A rapid chromosome mapping method for cloned fragments of yeast DNA. Genetics 105:857-872[Abstract/Free Full Text].

FELDMAN, H., M. AIGE, G. ALJINOVIC, B. ANDRE, and C. BACLET et al., 1994  Complete DNA sequence of yeast chromosome II. EMBO J. 13:5795-5809[Medline].

FISHEL, R., M. K. LESCOE, M. R. S. RAO, N. G. COPELAND, and N. A. JENKINS et al., 1993  The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75:1027-1038[Medline].

FRIEDBERG, E. C., 1985 DNA Repair. W. H. Freeman & Co., San Fran-cisco.

FRIEDBERG, E. C., 1990  The enzymology of DNA repair. Mutat. Res. 236:145-314[Medline].

GLICKMAN, B. W. and M. RADMAN, 1980  Escherichia coli mutator mutants deficient in methylation-instructed DNA mismatch correction. Proc. Natl. Acad. Sci. USA 77:1063-1067[Abstract/Free Full Text].

GORDENIN, D. A. and S. G. INGE-VECHTOMOV, 1981  Mechanisms of ultraviolet induced of ade2 mutants in diploids of yeast Saccharomyces Cerevisiae. Genetika (Moscow) 17:822-831[Medline].

GRACHEVA, L. M., T. A. EVSTUCHINA, S. V. KOVALTSOVA, and V. G. KOROLEV, 1992  The gene HIM1 of yeast Saccharomyces cerevisiae takes part in the correction of heteroduplex DNA. Genetika (Moscow) 27:56-65.

GRILLEY, M., J. HOLMES, B. YASHAR, and P. MODRICH, 1990  Mechanisms of DNA-mismatch correction. Mutat. Res 236:253-267[Medline].

HAN, H. J., A. YANAGISAWA, Y. KATO, J.-G. PARK, and Y. NAKAMURA, 1993  Genetic instability in pancreatic cancer and poorly differentiated types of gastric cancer. Cancer Res. 53:5087-5089[Abstract/Free Full Text].

HICKSON, I. D., H. M. ARTHUR, D. BRAMHILL, and P. T. EMMERSON, 1983  The E. coli uvrD gene product is DNA helicase II. Mol. Gen. Genet. 190:265-270[Medline].

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

HOLLINGSWORTH, N. M., 1995  MSH5, a novel MutS homolog, facilitates meiotic reciprocal recombination between homologs in Saccharomyces cerevisiae but not mismatch repair. Genes Dev. 9:1728-1739[Abstract/Free Full Text].

IONOV, Y., M. A. PEINADO, S. MALKHOSYAN, D. SHIBATA, and M. PERUCHO, 1993  Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature 363:558-561[Medline].

ITO, H., Y. FUKUDA, K. MURATA, and A. KIMURA, 1983  Transforma-tion of intact yeast cells treated with alkali cations. J. Bacteriol. 153:163-168[Abstract/Free Full Text].

IVANOV, E. L., S. V. KOVALTZOVA, and V. G. KOROLEV, 1989  Saccharomyces cerevisiae mutants with enhanced induced mutation and altered mitotic gene conversion. Mutat. Res. 213:105-115[Medline].

IVANOV, E. L., I. V. FEDOROVA, and S. V. KOVALTSOVA, 1992  Isolation and characterization of new mutants with enhanced spontaneous mutability in the yeast Saccharomyces cerevisiae. Genetika (Moscow) 27:47-55.

KOVALTZOVA, S. V. and V. G. KOROLEV, 1989  Influence of HIM mutations characterized by enhanced induced mutagenesis on spontaneous mitotic gene conversion in the ADE2 gene of yeast Saccharomyces cerevisiae. Genetika (Moscow) 25:2111-2120[Medline].

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

KRAMER, B., W. KRAMER, M. S. WILLIAMSON, and S. FOGEL, 1989b  Heteroduplex DNA correction in Saccharomyces cerevisiae is mismatch specific and requires functional PMS genes. Mol. Cell Biol. 9:4432-4440[Abstract/Free Full Text].

LEA, D. E. and C. A. COULSON, 1949  The distribution of the numbers of mutants in bacterial populations. J. Genet. 49:264-284.

LEACH, F. S., N. C. NICOLAIDES, N. PAPADOPOULOS, B. LU, and J. JEN et al., 1993  Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 75:1215-1225[Medline].

LI, G.-M. and P. MODRICH, 1995  Resotration of mismatch repair to nuclear extracts of H6 colorectal tumor cells by a heterodimer or human MutL homologs. Proc. Natl. Acad. Sci. USA 92:1950-1954[Abstract/Free Full Text].

LINDBLOM, A., P. TANNERGARD, B. WERELIUS, and M. NORDENSKJOLD, 1993  Genetic mapping of a second locus predisposing to hereditary non-polyposis colon cancer. Nature Genet. 5:279-282[Medline].

MA, C. and R. K. MORTIMER, 1983  Empirical equation that can be used to determine genetic map distances from tetrad data. Mol. Cell Biol. 3:1886-1887[Abstract/Free Full Text].

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

MARCIL, R. and R. D. HIGGINS, 1991  Direct transfer of plasmid DNA from yeast to E. coli by electroporation. Nucleic Acids Res. 20:917-918[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].

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

NEW, L., K. LIU, and G. F. CROUSE, 1993  The yeast gene MSH3 defines a new class of eukaryotic MutS homologoues. Mol. Gene. Genet. 239:97-108.

NICOLAIDES, N. C., N. PAPADOPOULOS, B. LIU, Y. WEI, and K. C. CARTER et al., 1994  Mutation of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371:75-80[Medline].

NOSKOV, V. N., K. NEGICHI, A. ONO, B. ONO, and H. HAUATSU, 1994  Mutagenicity of 5-bromouracil and N⁶-hydroxyadenine studied by yeast oligonucleotide transformation assay. Mutat. Res. 308:43-51[Medline].

PAPADOPOULOS, N., N. C. NICOLAIDES, Y.-F. WEI, S. M. RUBEN, and K. C. CARTER et al., 1994  Mutation of a mutL homolog in hereditary colon cancer. Science 263:1625-1628[Abstract/Free Full Text].

PAVLOV, YU. I., V. N. NOSKOV, YU. O. CHERNOFF, and D. A. GORDENIN, 1988  LYS2 gene mutability in yeast Saccharomyces cerevisiae of different ploidy. 2. Mutants induced by 6-N-hydroxylaminopurine and propiolactone. Genetika (Moscow) 24:1752-1760[Medline].

PAVLOV, YU. I., V. N. NOSKOV, E. K. LANGE, E. V. MOISEEVA, and M. R. PSHENICHNOV et al., 1991  The genetic activity of N⁶-hydroxyadenine and 2-amino-N⁶-hydroxyadenine in Escherichia coli, Salmonella typhimurium and Saccharomyces cerevisiae. Mutat. Res. 253:33-46[Medline].

PERKINS, D. D., 1949  Biochemical mutants of the smut fungus Ustilago maydis. Genetics 34:607-626[Free Full Text].

PROLLA, T. A., D.-M. CHRISTIE, and R. M. LISKAY, 1994  Dual requirement in yeast DNA mismatch repair for MLH1 and PMS1, two homologs of the bacterial mutl gene. Mol. Cell. Biol. 14:407-415[Abstract/Free Full Text].

RADMAN, M., 1988 Mismatch repair and genetic recombination, pp. 169–192 in Genetic Recombination, edited by P. KUCHERLAPATI and G. R. SMITH. American Society of Microbiologists, Washington, DC.

REENAN, R. A. G. and R. D. KOLODNER, 1992a  Isolation and characterization of two Saccharomyces cerevisiae genes encoding homologs of the bacterial HexA and MutS mismatch repair proteins. Genetics 132:963-973[Abstract].

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

ROMAN, H., 1956  A system selective for mutations affecting the synthesis of adenine in yeast. Compt. Rend. Trav. Lab. Carlsberg. Ser. Physiol. 26:299-314.

ROSS-MACDONALD, P. and G. S. ROEDER, 1994  Mutation of a meiosis-specific MutS homolog decreases crossing over but not mismatch correction. Cell 79:1069-1080[Medline].

SANCAR, A. and J. E. HEARST, 1993  Molecular matchmakers. Science 259:1415-1420[Abstract/Free Full Text].

SHCHERBAKOVA, P. V. and Y. I. PAVLOV, 1993  Mutagenic specificity of the base analog 6-N-hydroxylaminopurine in the URA3 gene of the yeast Saccharomyces cerevisiae. Mutagenesis 8:417-421[Abstract/Free Full Text].

STRAND, M., M. C. EARLEY, G. F. CROUSE, and T. D. PETES, 1995  Mutations in the MSH3 gene preferentially lead to deletions within tracts of simple repetitive DNA in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 92:10418-10421[Abstract/Free Full Text].

SU, S.-S. and P. MODRICH, 1986  Escherichia coli mutS-encoded protein binds to mismatched DNA base pairs. Proc. Natl. Acad. Sci. USA 83:5057-5061[Abstract/Free Full Text].

SU, S.-S., R. S. LAHUE, K. G. AU, and P. MODRICH, 1988  Mispair specificity of methyl-directed DNA mismatch correction. J. Biol. Chem. 263:6829-6835[Abstract/Free Full Text].

THIBODEAU, S. N., G. BREN, and D. SCHAID, 1993  Microsatellite instability in cancer of the proximal colon. Science 260:816-819[Abstract/Free Full Text].

VON BORSTEL, R. C., 1978  Measuring spontaneous mutation rates in yeast. Appendix: N. N. KHROMOV-BORISOV, Biometrical aspects of measuring mutation rates. Methods Cell Biol. 20:20-24.

WELSH, K. M., A.-L. LU, S. CLARK, and P. MODRICH, 1987  Isolation and characterization of the Escherichia coli mutH gene product. 1987 J. Biol. Chem. 282:15624-15629.

WILLIAMSON, M. S., J. C. GAME, and S. FOGEL, 1985  Meiotic gene conversion mutants in Saccharomyses cerevisiae. I. Isolation and characterization of pms1-1 and pms1-2. Genetics 110:609-646[Abstract/Free Full Text].

ZAKHAROV, I. A., S. A. KOZHIN, T. A. KOZHINA and I. V. FEDOROVA, 1984 Methods in Yeast Saccharomyces cerevisiae Genetics. Nauka, Leningrad.

This article has been cited by other articles:

				P. J. White, R. H. Borts, and M. C. Hirst Stability of the Human Fragile X (CGG)n Triplet Repeat Array in Saccharomyces cerevisiae Deficient in Aspects of DNA Metabolism Mol. Cell. Biol., August 1, 1999; 19(8): 5675 - 5684. [Abstract] [Full Text] [PDF]