Functional Interaction of MutY Homolog with Proliferating Cell Nuclear Antigen in Fission Yeast, Schizosaccharomyces pombe *

The MutY homolog (MYH) is responsible for removing adenines misincorporated on a template DNA strand containing G or 7,8-dihydro-8-oxoguanine (8-oxoG) and thus preventing G:C to T:A mutations. Human MYH has been shown to interact physically with human proliferating cell nuclear antigen (hPCNA). Here, we report that a similar interaction between SpMYH and SpPCNA occurs in the fission yeast Schizosaccharomyces pombe. Binding of SpMYH to SpPCNA was not observed when phenylalanine 444 in the PCNA binding motif of SpMYH was replaced with alanine. The F444A mutant of SpMYH expressed in yeast cells had normal adenine glycosylase and DNA binding activities. However, expression of this mutant form of SpMYH in aSpMYHΔ cell could not reduce the mutation frequency of the cell to the normal level. Moreover, SpMYH interacted with hPCNA, and SpPCNA interacted with hMYH but not with F518A/F519A mutant hMYH containing mutations in its PCNA binding motif. Although theSpMYHΔ cells expressing hMYH had partially reduced mutation frequency, the F518A/F519A mutant hMYH could not reduce the mutation frequency of SpMYHΔ cells. Thus, the interaction between SpMYH and SpPCNA is important for SpMYH biological function in mutation avoidance.

The MutY homolog (MYH) is responsible for removing adenines misincorporated on a template DNA strand containing G or 7,8-dihydro-8-oxoguanine (8-oxoG) and thus preventing G:C to T:A mutations. Human MYH has been shown to interact physically with human proliferating cell nuclear antigen (hPCNA). Here, we report that a similar interaction between SpMYH and SpPCNA occurs in the fission yeast Schizosaccharomyces pombe. Binding of SpMYH to SpPCNA was not observed when phenylalanine 444 in the PCNA binding motif of SpMYH was replaced with alanine. The F444A mutant of SpMYH expressed in yeast cells had normal adenine glycosylase and DNA binding activities. However, expression of this mutant form of SpMYH in a SpMYH⌬ cell could not reduce the mutation frequency of the cell to the normal level. Moreover, SpMYH interacted with hPCNA, and SpPCNA interacted with hMYH but not with F518A/ F519A mutant hMYH containing mutations in its PCNA binding motif. Although the SpMYH⌬ cells expressing hMYH had partially reduced mutation frequency, the F518A/F519A mutant hMYH could not reduce the mutation frequency of SpMYH⌬ cells. Thus, the interaction between SpMYH and SpPCNA is important for SpMYH biological function in mutation avoidance.
Oxidation damage to DNA can induce mutagenesis and lead to degenerative diseases. One of the most stable products of DNA damage resulting from reactive oxygen species is 7,8dihydro-8-oxoguanine (8-oxoG 1 or GO). If not repaired, GO lesions in DNA can produce A/GO mismatches during DNA replication (1) and result in G:C to T:A transversions (2)(3)(4)(5). In Escherichia coli, MutT, MutM, and MutY reduce the mutagenic effects of GO lesions (6,7). MutT hydrolyzes oxidized dGTP and depletes it from the nucleotide pool (8). The function of the MutM (Fpg) protein is to remove the mutagenic GO from C/GO adducts (9). MutY adenine glycosylase is responsible for correcting A/GO as well as A/G and A/C mismatches (10 -12). Thus, MutY provides a measure of defense by removing ad-enines misincorporated opposite GO or G following DNA replication (6,13,14).
The DNA repair pathways that protect the cells from the mutagenic effects of GO are highly conserved. Homologs of MutT, MutM, and MutY have been identified in humans (15)(16)(17)(18)(19)(20)(21). These three enzymes (hMTH1, hOGG1, and hMYH), like their E. coli homologs, are proposed to function in the reduction of 8-oxoG in the human genome (22). Although no mutY homolog has been found in the budding yeast Saccharomyces cerevisiae, a mutY homologous (SpMYH) gene of fission yeast Schizosaccharomyces pombe was identified and cloned (23). The SpMYH gene encodes a 461-amino acid protein, which displays 28 and 31% identity to E. coli MutY and hMYH, respectively. Expression of SpMYH cDNA in an E. coli mutY mutant cell is able to reduce the mutation frequency. As with the E. coli MutY, the SpMYH protein has both adenine glycosylase and weak AP lyase activities on A/G and A/GO mismatches (23). An SpMYH knockout strain (SpMYH⌬) of S. pombe has been shown to be a mutator (24). Disruption of SpMYH also causes increased sensitivity to H 2 O 2 but not to UV irradiation. Thus, MutY homolog plays an important role in defense against oxidative stress in eukaryotes.
We have shown that hMYH is directly associated with human apurinic/apyrimidinic endonuclease (hAPE1), proliferating cell nuclear antigen (hPCNA), and replication protein A (hRPA), suggesting that hMYH plays a role in the long patch base excision repair pathway (25). It has been suggested that hMYH repair is coupled to DNA replication through docking with hPCNA and hRPA (25,26). The coupling to DNA replication may provide a signal to target the MYH repair to the daughter DNA strands. In such a model, MYH can remove adenines on the daughter strands mismatched with guanines or 8-oxoG as a result of DNA replication errors but cannot excise the adenines on the template strands. Here, we provide direct evidence that the interaction between SpMYH and SpPCNA of S. pombe is important for SpMYH biological function in mutation avoidance. A mutant form of SpMYH, which has normal glycosylase activity but cannot interact with Sp-PCNA, is partially defective in vivo. In addition, interactions between MYH and PCNA proteins from both S. pombe and humans are interchangeable. S. pombe will be an excellent model system to study mammalian MYH repair.

EXPERIMENTAL PROCEDURES
Construction of Mutant SpMYH, Wild-type hMYH, and Mutant hMYH in a Yeast Expression Vector-The cDNA of hMYH was cloned into the S. pombe expression vector pREP41X (American Type Culture Collection) by the PCR method. XHO-5-MYH and XMA-3-MYH were used as primers, and pET11-hMYH7 (27) was used as the template to amplify the cDNA by Pfu DNA polymerase (Stratagene, La Jolla, CA) of hMYH gene. The cDNA of SpMYH containing the F444A mutation and cDNA of hMYH containing the double mutation F518A/F519A were constructed by the PCR method. 5Ј Primer (XHO-5-SP) and mutation primer (F444A) were used as primers, and pSPMYH19 (23) was used as the template to amplify the cDNA of SpMYH gene containing a F444A mutation. Similarly, XHO-5-MYH and XMA-3-AA were used as primers, and pET11-hMYH7 (27) was used as the template to amplify the cDNA of hMYH gene containing a double mutation of F518A/F519A. All the oligonucleotides used are listed in Table I. The three resulting PCR products were digested with XhoI and XbaI, cloned into pREP41X, and transformed into SpMYH knockout cells, JSP303-Y4 (SpMYH⌬) (24), by electroporation. The expression vector pREP41X contains the nmt1 promoter that can be regulated by varying concentrations of thiamine. Transformed cells were selected with Leu ϩ phenotype on the EMM agar plates. One clone containing hMYH gene (pREPHY-WT) and two clones containing mutations in the SpMYH and hMYH genes, pREPSPY-A and pREPHY-AA, respectively, were confirmed by DNA sequencing.
The His-tagged SpPCNA protein was obtained by cloning the Sp-PCNA gene into vectors pET21a and pET28b (Novagen, Inc., Madison, WI). The SpPCNA gene was amplified from a S. pombe cDNA library in pGADGH (kindly provided by D. Beach, Cold Spring Harbor Laboratory). To clone into pET21a, PCR products were produced by primers PCNA-5-NDE and PCNA-3-ECO, digested with NdeI and EcoRI, and ligated into the NdeI-EcoRI-digested vector. To clone into pET28b, PCR products were produced by primers PCNA-5-NDE and PCNA-3-BAM, digested with NdeI and BamHI, and ligated into the NdeI-BamHIdigested vector. Two clones containing the SpPCNA gene (pET21-SPCNA and pET28-SPCNA) were confirmed by DNA sequencing.
Expression and Purification of the Recombinant Proteins-CHANG219 and CHANG220 were used as primers, and pREPSPY-A was used as a template to amplify SpMYH cDNA containing the F444A mutation. The PCR product was cleaved by NdeI and BamHI and ligated into pET28b to obtain the clone pET28SPY-A. E. coli BL21-star cells (Stratagene) harboring the expression plasmid, pET28SPY-A, were grown in LB broth containing 100 mg/ml kanamycin at 25°C. Protein expression was induced at an A 590 of 0.6 by the addition of isopropyl-1-thio-␤-D-galactoside to a final concentration of 0.4 mM, and the cells were harvested 16 h later. The cell paste, from a 500-ml culture, was resuspended in 8 ml of buffer A (20 mM potassium phosphate, pH 7.4, 50 mM NaCl, 1 mM dithiothreitol, 0.2 mM phenylmethylsulfonyl fluoride). The His-tagged SpMYH(F444A) mutant protein was purified by a nickel-agarose column (Qiagen Inc., Valencia, CA) according to the manufacturer's protocol, and the His-tag was removed by thrombin (Novagen, Inc., Madison, WI) cleavage.
SpPCNA protein was purified as a His-tagged protein from the BL21-star cell harboring plasmid pET28-SPCNA, and then the tag was removed similar to the procedures described above for SpMYH(F444A) mutant protein. Recombinant SpMYH expressed in E. coli was purified according to the procedures described by Lu and Fawcett (23). Human PCNA expressed in E. coli was from Dr. Mike O'Donell (Rockfeller University and Howard Hughes Medical Institute).
Nickel-agarose Affinity Binding-The His-tagged SpPCNA from the BL21-star cell harboring plasmid pET21-SPCNA was bound to nickelagarose (Qiagen Inc.) according to the manufacturer's procedures. Purified SpMYH and SpMYH(F444A) (200 ng) expressed in E. coli were added to beads and incubated at 4°C for 1 h. After washing with buffer N (50 mM potassium phosphate, pH 8.0, 300 mM NaCl) containing 50 mM imidazole, the bound proteins were eluted by buffer N containing 250 mM imidazole. The unbound and eluting fractions were fractionated by 10% SDS-polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. The affinity-purified SpMYH polyclonal antibodies (28) were used for Western blotting analysis.
Assays of SpMYH Binding and Glycosylase Activities in Yeast Extracts-The binding and glycosylase assays for SpMYH in yeast cell extracts with an A/GO-containing DNA were described previously (23, 24) except using 9 mM EDTA. The DNA substrate was a 20-mer duplex DNA containing an A/8-oxoG mismatch (see Table I) that was labeled at the 3Ј-end of the mismatched A-containing strand.
Measurement of Mutation Frequency-Five independent yeast colonies were grown to late log phase in EMM containing 0.1 mg/ml uracil. Additional amino acids were supplemented for the wild-type strain (0.1 mg/ml Leu and His) and the SpMYH⌬ strain (0.1 mg/ml Leu). Each culture (0.2 ml) was plated onto EMM agar plates containing 1 mg/ml 5-fluoro-orotic acid (FOA) and 0.1 mg/ml uracil. FOA-resistant colonies were counted after 5 days of growth. The cell titer was determined by plating 0.1 ml of a 10 Ϫ4 dilution onto plates without FOA. The mutation frequency was calculated as the ratio of FOA-resistant cells to the total cells. The measurement was repeated more than three times.

RESULTS AND DISCUSSION
SpMYH Physically Interacts with SpPCNA-It has been shown that PCNA can interact with many proteins involved in DNA replication and repair and that these PCNA-binding proteins share a common motif (30,31). Generally, this motif contains a glutamine (Gln) at position 1, an aliphatic residue such as leucine (Leu), isoleucine (Ile), or methionine (Met) at position 4, and a pair of aromatic residues (Phe or Tyr) at

5Ј-GGAGCTGGATCCCTACTCCTCATCCTCCTCACC-3Ј
His-SpPCNA in pET28b CHANG143 positions 7 and 8. The residues flanking the conserved motif also show a preponderance of proline and basic residues. The conserved PCNA binding motif QXXLXXFF is found in human and mouse MutY homologs (20,32) (Fig. 1A). Parker et al. (25) have shown that hMYH is directly associated with hPCNA and Boldogh et al. (26) have shown that hMYH colocalizes with hPCNA to replication foci. The hPCNA binding activity is located at the C terminus of hMYH containing residues 505-527 (25).
Although S. pombe MYH (23) contains the conserved leucine (Leu-441) at position 4 as well as a phenylalanine (Phe-444) at position 7 within the PCNA binding motif, it does not contain either a glutamine at position 1 or a phenylalanine at position 8 (31) (Fig. 1A). To demonstrate that the same MYH-PCNA interaction occurs in the yeast system, we performed affinity binding experiments with full-length His-tagged SpPCNA and purified SpMYH. As shown in Fig. 1B, SpMYH was detected in the bound fraction of nickel-agarose containing bound SpPCNA (Fig. 1B, lanes 1 and 2) but not in the nickel-agarose containing bovine serum albumin (Fig. 1B, lanes 3 and 4). When Phe-444 in the conserved SpPCNA binding motif was mutated to Ala, SpMYH could not bind SpPCNA (Fig. 1B, lanes 5 and 6). Thus, there is a direct association between SpMYH and SpPCNA.
The results indicate that glutamine at position 1 and phenylalanine at position 8 are dispensable for SpMYH and SpPCNA interaction. Several PCNA-binding proteins also lack this conserved glutamine (30) although this residue has been reported as being essential for the interaction between 5Ј-methylcytosine DNA methyltransferase and PCNA (33). One factor to enhance the SpMYH-SpPCNA interaction may be the presence of clusters of basic residues flanking the conserved motif of SpMYH (Fig. 1A).

The Interactions between MYH and PCNA Proteins from Both S. pombe and Humans Are Interchangeable-Because
SpMYH and hMYH have different PCNA binding motifs (Fig.  1A), we speculate whether SpMYH can bind to hPCNA or SpPCNA can bind to hMYH. The affinity binding experiments with GST fusion proteins of SpMYH and hMYH were performed with purified hPCNA and SpPCNA, respectively. Human PCNA was detected in the GST-SpMYH pellets ( Fig. 2A,  lanes 1 and 2) but not in the GST beads ( Fig. 2A, lanes 3 and 4). Conversely, SpPCNA was detected in the pellets of ⌬231-hMYH fused to GST (Fig. 2B, lanes 1 and 2) but not in the GST beads (Fig. 2B, lanes 5 and 6). Residues Phe-518 and Phe-519 of hMYH have been identified as being essential for hPCNA binding (25). When both Phe-518 and Phe-519 of hMYH were mutated to Ala residues, GST-hMYH could not bind SpPCNA (Fig. 2B, lanes 3 and 4). Thus, the interactions between MYH and PCNA proteins from both S. pombe and humans are interchangeable.
The physical interactions between MYH and PCNA proteins from both S. pombe and humans are consistent with the high homologies of these proteins. The PCNA proteins from both organisms have 51% identity whereas SpMYH displays 31% identity to hMYH. The anti-hPCNA antibody directed toward residues 112-121 can also cross-react with SpPCNA (Fig. 2B). Several proteins have been shown to bind to a hydrophobic pocket consisting of the interdomain connector loop (residues 118 -135) and loops on the C termini of the PCNA trimer (34,35). A ClustalW alignment indicates that amino acid sequences in this pocket are very conserved between SpPCNA and hPCNA.
F444A Mutant of SpMYH Had Normal Adenine Glycosylase and DNA Binding Activities-The SpMYH protein is highly homologous to E. coli MutY and hMYH. The N-terminal domain of E. coli MutY has the catalytic activity (36 -40); however, the C-terminal domain is important for GO recognition (36,37,40,41). The PCNA binding motifs located at the Cterminal ends of eukaryotic MYH are not conserved in bacterial MutY proteins. To test that Phe-444 is not at the active site of SpMYH, mutant SpMYH (F444A) was generated and expressed in the SpMYH⌬ yeast cell. Because wild-type and F444A mutant SpMYH from plasmids were expressed about 30 -60-fold higher than that from chromosomal gene (Fig. 3,  lanes 1, 3, and 5), different amounts of cell extracts were used in the SpMYH assays. When the extract from the SpMYH (F444A) mutant was assayed for SpMYH activities, the glycosylase (Fig. 4A, lane 4) and DNA binding (Fig. 4B, lane 4) activities of SpMYH with an A/GO-containing DNA substrate could be detected. The glycosylase activity of F444A mutant protein was similar to that of the expressed wild-type SpMYH (Fig. 4A, compare lanes 3 and 4) because the F444A mutant was slightly higher expressed in yeast cells than that of wildtype enzyme (Fig. 3, compare lanes 3 and 5). The substrate binding activity of expressed F444A mutant was slightly weaker than that of the expressed wild-type enzyme (Fig. 4B,  compare lanes 3 and 4) but was higher than that of the SpMYH from the chromosomal gene (Fig. 4B, compare lanes 1 and 4). Thus, Phe-444 of SpMYH is not involved in the catalytic activity but is essential for binding to SpPCNA.
Expression of F444A Mutant of SpMYH in the SpMYH⌬ Cells Is a Mutator-We have shown that a SpMYH⌬ S. pombe strain, JSP303-Y4 (SpMYH⌬), is a mutator (24). Expression of wild type SpMYH in the SpMYH⌬ can reduce the mutation frequency to the same level as wild-type cells. Because the nmt1 promoter controls the expression of SpMYH cDNA in pREP41X, SpMYH protein expression can be regulated by varying concentrations of thiamine in the minimal medium. At 5 g/ml thiamine, the expression of SpMYH is almost completely suppressed (24) (Fig. 3, lane 4). Therefore, cells growing in minimal media containing 5 g/ml thiamine provide good controls for mutation frequency measurement. As shown in Table II, the SpMYH⌬ cell expressing wild-type SpMYH protein in media without thiamine had a mutation frequency similar to that of the wild type (compare lines 1 and 3) although the expression level of SpMYH from plasmid is about 30-fold higher than that of the SpMYH from chromosomal gene (Fig. 3,  lanes 1 and 3). It appears that the SpMYH protein amount from the chromosomal gene is sufficient to maintain the genome stability. When the expression of SpMYH was inhibited by 5 g/ml thiamine, the cell's mutation frequency was much higher than that of the wild type (Table II, compare lines 1 and 4).
To test whether interaction with SpPCNA is important for the SpMYH function in vivo, SpMYH⌬ yeast cell expressing mutant SpMYH (F444A) was tested for mutation frequency. The expression level of F444A mutant protein in yeast cells was slightly higher than that of wild-type enzyme in the absence of thiamine (Fig. 3, compare lanes 3 and 5). The mutation frequencies of SpMYH⌬ yeast cells expressing mutant SpMYH were about 25-fold higher than that of the wild type and was slightly lower than that of the parental SpMYH⌬ strain (Table  II, compare lines 5 and 6 with lines 1 and 2). Thus, the F444A mutant SpMYH could not complement the chromosomal Sp-MYH mutation. Therefore, a mutant SpMYH that retains substrate binding and glycosylase activities but cannot interact with SpPCNA is nearly defective on oxidative DNA repair. These results provide direct evidence that the interaction be-  1 and 2) or GST alone (lanes 3 and 4) were immobilized on beads and bound to purified hPCNA. Western blot was probed with antibody against hPCNA. S, supernatant; P, pellet. B, SpPCNA physically interacts with hMYH but not with mutant hMYH(F518A/F519A). Lanes 1 and 2 used GST fusion protein containing residues 232-535 of hMYH (⌬231-hMYH), lanes 3 and 4 used GST fusion protein containing mutant (M) ⌬231-hMYH(F518A/F519A), and lanes 5 and 6 used GST alone. Western blotting to detect SpPCNA was performed with antibody against hPCNA due to their high homology. S, supernatant; P, pellet.
tween SpMYH and SpPCNA is important for SpMYH biological function in mutation avoidance.
As shown above, the physical interactions between both MYH and PCNA proteins from S. pombe and humans are interchangeable. We then expressed hMYH and mutant hMYH (F518A and F519A) in the SpMYH⌬ yeast cell and tested their in vivo complementation activities. When wildtype hMYH was expressed in the SpMYH⌬ cells, the mutation frequency was 8-fold higher than that of the wild type but was 5-fold lower than that of the parental SpMYH⌬ strain (Table II, compare lines 7 and 8 with lines 1 and 2). Therefore, hMYH is partially functional in S. pombe cells. However, when the F518A/F519A mutant hMYH was expressed in the SpMYH⌬ cells, the mutation frequency was the same as that of the parental SpMYH⌬ strain (Table II, compare last two lines with line 2). As shown in the Western blot in Fig. 3, the expression levels of both wild-type and mutant hMYH (lanes 9 and 10) from the plasmids were similar. Additionally, the expression levels of hMYH and SpMYH in yeast cells were comparable when they were compared to known amounts of recombinant hMYH (Fig. 3, compare lanes 9 and 10 with lane 11) and SpMYH (Fig. 3, compare lanes 3 and 5 with lanes 6 -8) expressed in E. coli, respectively. Therefore, partial rescue of the mutator phenotype of SpMYH⌬ cells by wild-type hMYH is not due to the low expression levels of hMYH. It is possible that hMYH may not act as SpMYH in the yeast cells because hMYH does not interact well with other enzymes involved in the SpMYH repair pathway such as AP endonuclease or single-stranded DNA-binding protein. It has been shown that mouse MYH activity can be stimulated by human AP endonuclease (32) and that hMYH physically interacts with hRPA (25).
It has been suggested that hMYH base excision repair is coupled to DNA replication through docking with hPCNA and hRPA (25,26). PCNA may be important to position MYH protein on the replication fork and may allow MYH to discriminate between the parental and daughter strands. In such mechanism, MYH excises misinserted A from template GO but does not act on template A when the DNA polymerase misinserts GO. However, some in vivo MYH function is retained in the F444A mutant SpMYH. Thus, the role of PCNA is not absolutely essential. This partial retention of the function of mutant SpMYH(F444A) may reflect that the mutant protein can still act on A/GO mismatches on daughter DNA strands through interaction with other replication proteins such as RPA. It is also possible that DNA polymerase has a higher frequency of inserting A on the GO template than inserting GO on the A template.
The PCNA sliding clamp also interacts with several DNA repair proteins. When Phe residues at the PCNA binding motifs of Fen1 and DNA ligase I are substituted by Ala, their functions in long patch base excision repair are reduced (42,43). Several mismatch repair proteins also interact with PCNA (44 -47), and the conserved PCNA binding motif can be found in the MSH3 and MSH6 sequences. The interactions of yeast Msh3 and Msh6 with PCNA have been shown to be important for mismatch repair (44,45). PCNA may enhance early steps in mismatch repair or strand discrimination (44). However, p21 SDI1/WAF1/CIP1 -mediated inhibition of DNA replication is not dependent on its PCNA binding (48). It has been suggested that PCNA may act as a molecular adaptor, coordinating and regulating the actions of DNA replication, DNA repair, and cell cycle control. However, the mechanism by which PCNA selects the appropriate partners remains unclear.