Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Biochemistry and genetics of eukaryotic mismatch repair.

Biochemistry and genetics of eukaryotic mismatch repair. Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press REVIEW Biochemistry and genetics of eukaryotic mismatch repair Richard Kolodner Charles A. Dana Division of Human Cancer Genetics, Dana-Farber Cancer Institute, Boston, Massachusetts 021 15 USA and Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 021 15 USA The process of mismatch repair was first postulated to match repair, base-specific mismatch repair systems, explain the results of experiments on genetic recombi- and cancer genetics, see other recent reviews (Modrich nation and bacterial mutagenesis. Mismatch repair has 1991; Eshleman and Markowitz 1995; Fishel and Kolod- long been known to play a major role in two cellular ner 1995; Friedberg et al. 1995; Kolodner 1995; Marra processes: (1) the repair of errors made during DNA rep- and Boland 1995; Modrich and Lahue 1996). lication or as the result of some types of chemical dam- age to DNA and DNA precursors; and (2) the processing The E. coli MutHLS mismatch repair pathway of recombination intermediates to yield new configura- tions of genetic markers. More recent studies have sug- The E. coli MutHLS mismatch repair pathway is a gen- gested that mismatch repair may also be crucial for (1) eral DNA repair pathway that recognizes and repairs all the regulation of recombination events between diver- single-base mispairs except C.C (Modrich 199 1 Fishel gent DNA sequences that could result in different types and Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; of genetic instability (Rayssiguier et al. 1989; Selva et al. Modrich and Lahue 1996). It also repairs small insertion1 1995; Datta et al. 1996), (2) some types of nucleotide deletion mispairs, although it may not efficiently recog- excision repair responsible for repair of physicallchemi- insertionldeletion mispairs that have more nize most cal damage to DNA (Karran and Marinus 1982; Fram et than 4 unpaired bases. The basic repair reaction cata- al. 1985; Feng et al. 1991; Mellon and Champe 1996), and lyzed by this pathway is understood in considerable de- (3) participating in a cell-cycle checkpoint control sys- tail because it has been reconstituted in vitro with DNA tem by recognizing certain types of DNA damage and substrates containing mispaired bases, MutH, MutL, triggering cell-cycle arrest or other responses to DNA MutS, and UvrD (helicase 11) proteins, DNA polymerase damage (Hawn et al. 1995; Anthoney et al. 1996). I11 holoenzyme, DNA ligase, single-strand DNA-binding The most extensively characterized general mismatch protein (SSB), and any one of the single-stranded DNA repair system is the Escherichia coli MutHLS system, exonucleases-Exo I, Exo VII, and RecJ protein (Fig. 1; which repairs a broad spectrum of mispaired bases and Lahue et al. 1989; Modrich 1991; Grilley et al. 1993; has been reconstituted with purified enzymes. Eukary- Fishel and Kolodner 1995; Friedberg et al. 1995; Kolod- otes are known to contain a mismatch repair system that ner 1995; Modrich and Lahue 1996). The reaction in- has at least some components that are highly related to volves mismatch-dependent nicking of the unmethy- key components of the bacterial MutHLS mismatch re- lated strand at a hemimethylated GATC site and degra- pair system. The observation that defects in mismatch dation from the nick past the mismatch followed by repair genes are linked to both inherited cancer suscep- resynthesis. The roles of many of the proteins that func- tibility and some sporadic cancers has generated consid- tion in this reaction have been elucidated. The MutS erable interest in the gene products that function in eu- protein binds to DNA at the site of a mispaired base and karyotic mismatch repair. The goal of this review is to is responsible for mismatch recognition. No activity has discuss recent studies on the mechanisms of MutHLS- been assigned to MutL, although it interacts with MutS like mismatch repair in the yeast Saccharomyces cere- bound to a mispaired base and is required for activation visiae and in humans and to relate insights derived from of MutH. MutH is an endonuclease that nicks hemime- these studies to human cancer genetics. Given space thylated DNA on the unmethylated strand when acti- constraints, it is difficult to cover everything known vated by MutS and MutL in the presence of a mismatch. about mismatch repair or to reference all of the relevant The requirement for MutH in the reaction can be re- work that has been done in this area. However, a brief placed by a pre-existing nick in the DNA. The repair overview of the E. coli MutHLS pathway is presented reaction can utilize hemimethylated sites that are either below to allow comparison of the E. coli and eukaryotic 5' or 3' to the mispair. Excision requires UvrD (helicase mismatch repair pathways and proteins. For more de- 11) and one of the single-stranded DNA exonucleases- tailed information, particularly related to bacterial mis- Exo 1 (3' exonuclease), Exo VII (3' and 5' exonuclease), GENES & DEVELOPMENT 10:1433-1442 O 1996 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/96 $5.00 1433 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner vincing evidence has implicated DNA methylation in strand discrimination during mismatch repair in other eukaryotes (Modrich 1991; Fishel and Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; Modrich and Lahue ///A /- MutH 1996). Thus, it seems possible that the mechanism of strand discrimination in eukaryotes will prove to be quite different from that observed in E. coli. Exo I, Exo VII or RecJ, Helicase II, MutHLS-like mismatch repair in eukaryotes DNA Pol Ill, SSB and 3' DNA ligase Two lines of experimentation have indicated that eu- karyotes have a broad-spectrum mismatch repair system related to the bacterial MutHLS system. First, a series of genetic studies led to the identification of s.' cerevisiae homologs of the bacterial mutL and mutS gene products and the demonstration that these gene products were required for mismatch repair (Williamson et al. 1985; Figure 1. Illustration of the action of the E. coli MutHLS mis- Bishop et al. 1987; W. Kramer et al. 1989; Reenan and match repair system on a mispair at a replication fork. Repair is Kolodner 1992a; Reenan and Kolodner 1992b; Kolodner initiated by binding of MutS protein to a mismatch. The sub- 1995). Second, biochemical studies of higher eukaryotic sequent binding of MutL to MutS is required to activate MutH, which then nicks the unmethylated strand of DNA at hemim- cells have demonstrated in vitro nick-directed revair of a ethylated GATC sites. Nicking of the unmethylated strand is variety of different mispaired bases similar to the repair then followed by the excision from the nick to the mispair and reactions catalyzed by the E. coli MutHLS system (Fang resynthesis to fill in the resulting gap. These interactions result and Modrich 1993; Modrich and Lahue 1996). More re- in the coupling of mismatch repair to DNA replication, so that cently, biochemical studies have demonstrated that mismatches formed during DNA replication are repaired using these mismatch repair reactions are dependent on eu- the methylated parental strand as template, resulting in a re- karyotic homologs of the bacterial mutL and mutS gene duction of misincorporation errors. (Adapted from Kolodner products (Umar et al. 1994; Boyer et al. 1995; Drum- 1995.1 mond et al. 1995: Li and Modrich 19951. A considerable body of evidence indicates that this plays an important role in maintaining replication fidelity, pro- and RecJ protein (5' exonucleasej-depending on cessing recombination intermediates, and regulating re- whether the nicked, unmethylated site is 5' or 3' to the combination in response to sequence divergence. These mispair. Once excision has occurred, resynthesis is me- subjects, however, are beyond the scope of this review. diated by DNA polymerase I11 holoenzyme, SSB, and DNA ligase. Homologs of the bacterial MutS proteins A crucial feature of the MutHLS mismatch repair pathway is its ability to preferentially repair the daugh- A series of studies has demonstrated that there are at ter DNA strand after DNA synthesis, thus increasing the least six S. cerevisiae proteins, MSH1-MSH6, which fidelity of DNA replication. Normally, DNA in E. coli is show a high degree of amino acid similarity with the methylated at GATC sites by the Dam methylase. This Bacterial MutS proteins (Fig. 1; Table 1). Homologs of is a postreplication modification, however, and the many of these proteins have been identified in other or- daughter DNA strand is transiently unmethylated after ganisms and are the subject of intCnse study. As will be DNA replication. The MutHLS system utilizes this discussed below, there is evidence that three of these modification asymmetry to direct repair to the unmeth- proteins, MSH2, MSH3, and MSH6, function in a eu- ylated daughter strand via the activity of MutH, which karyotic MutHLS-like mismatch repair pathway. The in- nicks the unmethylated DNA strand at hemimethylated volvement of three MutS homologs in mismatch repair GATC sites. It is important to point out that the require- provides evidence for the relative complexity of eukary- ment for MutH and the use of Dam methvlation to dis- otic mismatch repair compared to the bacterial para- tinguish between newly replicated and parental DNA digm. appear to be unique to certain bacteria. Other bacteria It was orignally thought that the eukaryotic MSH2 like Streptococcus pneumoniae have a mismatch repair protein, a homolog of the bacterial MutS protein, was the system that is closely related to the MutHLS system but mismatch repair protein that recognizes mispaired bases the Streptococcus system, referred to as the Hex system, in DNA (Reenan and Kolodner 1992a,b). Purified MSH2 is not presently known to utilize a MutH homolog or from both S. cerevisiae and human cells recognizes both DNA methylation as a mechanism for strand discrimi- single-base mispairs and, with a higher affinity, multiple nation in the repair process (Modrich 1991; Fishel and base insertion/deletion mispairs (Fishel et al. 1994; Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; Alani et al. 1995). More recently, it has been appreciated Modrich and Lahue 1996). DNA methylation does not that human MSH2 copurifies with a second MutS ho- occur in S. cerevisiae (Proffitt et al. 1984), and no con- molog, human GTBPIpl60 (Drummond et al. 1995; Pal- 1434 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair Table 1. Names and alternate names for mismatch repair al. 1996). Mutations in MSH2 were found to cause a genes and proteins discussed strong general mutator phenotype resulting in the accu- mulation of both frameshift and single-base substitution E. coli protein S, cerevisiae protein Human protein mutations, high rates of frameshift mutation reversion, MutS MSH2 MSH2 and dinucleotide repeat instability, consistent with a MSH3 MSH3, DUGI, MRPl general role in mismatch repair (Reenan and Kolodner MSH6 GTBP, p 160 1992a; Strand et al. 1993; Marsischky et al. 1996). In MutL PMS 1 PMS2 contrast, mutations in MSH6 caused only a modest mu- MLHl MLHl tator phenotype that was confined primarily to the ac- MLH2 PMS 1 cumulation of single-base substitution mutations, sug- General nomenclature: Gene names, uppercase italics; muta- gesting that MSH6 might be required primarily for rec- tion names, lowercase italics; protein names, uppercase except ognition of single-base substitution mispairs and to a for E. coli poteins, where only the first letter of each word is lesser extent single base insertionldeletion mispairs capitalized. (Marsischky et al. 1996). Consistent with this latter Note on MutL-related proteins: Human PMS2 is more closely view, pms3 mutants, which had been shown previously related to S. cerevisiae PMSl than human PMSl is related to S. in transformation assays that detect mismatch repair di- cerevisiae PMS1. Human PMSl is more closely related to S. rectly to cause a defect in the repair of single-base sub- cerevisiae MLH2 (see Fig. 4). stitution mispairs but not in the repair of single-base References: E. coli MutS-related proteins: (Reenan and Kolod- insertionldeletion mispairs, are caused by mutations in ner 1992a, Fishel et al. 1993; Leach et al. 1993; New et al. 1993; Drummond et al. 1995; Palombo et al. 1995; Marsischky et al. MSH6 (B. Kramer et al. 1989; Marsischky et al. 1996). 1996; Watanabe et al. 1996). E. coli MutL-related proteins: Also consistent with this view is the observation that (Bronner et al. 1994; Nicolaides et al. 1994; Papadopoulos et al. msh6 mutations cause only a small increase in dinucle- 1994; Prolla 1994a,b). otide repeat instability (Johnson et al. 1996). Unlike either msh2 or msh6 mutants, msh3 mutants have little if any mutator phenotype in forward mutation ombo et al. 1995), which appears to be the homolog of S. and frameshift reversion assays and show increased di- cerevisiae MSH6 (Figs. 2 and 3, and below). The obser- nucleotide repeat instability, albeit at lower levels, than vation that the heterodimer of human MSH2 and GTBP that seen in msh2 mutants (New et al. 1993; Alani et al. (MSH6) recognizes mispaired bases initially suggested 1994; Strand et al. 1995; Marsischky et al. 1996;). Strik- that the entire complex could function as the MutS ingly, when double mutant strains were analyzed, msh3 equivalent in eukaryotic mismatch repair (Drummond and msh6 mutations showed a large synergistic effect on et al. 1995; Palombo et al. 1995). The available data on the rate of accumulation of mutations in frameshift re- mismatch repair-defective tumor cell lines, however, version assays that measure the formation and repair of suggest that the role of a MSH2-GTBP (MSH6) complex single-base insertion1 deletion mispairs (Marsischky et in mismatch recognition may not be simple. In vivo, al. 1996). A similar synergistic effect was seen in forward tumor cell lines lacking MSH2 appear to have a high degree of microsatellite instability at both dinucleotide and mononucleotide repeat loci (Bhattacharyya et al. Mouse 1994; Shibata et al. 1994; Umar et al. 1994b; Boyer et al. Rat Human 1995; Papadopoulos et al. 1995). In contrast, tumor cell Xenopus lines lacking GTBP (MSH6) appear to have less pro- Drosophila S.cerevisiae nounced microsatellite instability: Dinucleotide repeat Human instability is difficult to detect in such cell lines, and Mouse Sxerevisiae mononucleotide repeat instability is less pronounced S.cerevisiae compared to msh2 mutant lines (Bhattacharyya et al. Human E.coli 1994; Shibata et al. 1994; Papadopoulos et al. 1995). Bio- MUTS MUTS S.typhirnurium chemical analysis also suggests that mutations in MSH2 MUTS H.influenzae and GTBP (MSH6) have somewhat different effects on MUTS A.vinlandii MUTS T.aquaticus mismatch repair in in vitro mismatch repair assays MUTS T.thermophilus (Drummond et al. 1995). Finally, mutations in MSH2 are MUTS B.subtilis HEXA S.pneurnoniae common in hereditary nonpolyposis colorectal cancer MSHI S.cerevisiae (HNPCC) families, whereas mutations in GTBP (MSH6) -I I 'J-L;:;3;::: have not been found (Papadopoulos et al. 1995; Liu et al. 1996). These data suggest that MSH2 and GTBP (MSH6) are not genetically equivalent. Genetic and biochemical analyses of mismatch repair Figure 2. Family tree of known MutS homolog proteins. All of in S. cerevisiae have suggested that mismatch recogni- the sequences used for this construction were either retrieved tion involves three MutS homologs: MSH2, MSH3, and from GenBank or are from unpublished studies; human MSH5, MSH6 (the homolog of human GTBPIpl GO), which form N. Winand and R. Kolodner (unpubl.); mouse MSH6, G. Crouse two different heterodimeric complexes (Marsischky et and R. Kolodner (unpubl.). GENES & DEVELOPMENT 1435 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press MSH2-MSH3 complexes about equally, and that repair Single-base mispair recognition of larger insertionldeletion mispairs likely utilizes the MSH2-MSH3 complex 5-10 times as frequently as the MSH2-MSH6 complex (Johnson et al. 1996; Marsischky et al. 1996). A crucial test of this model will require the purification of these two complexes and the demonstra- tion that they have the required mispair recognition properties. The genetics of S. cerevisiae MSH2, MSH3, and MSH6, Insertion-deletion mispair recognition where comparable data exist, are consistent with the re- sults of the analysis of human tumor cell lines. Like S. cerevisiae msh2 mutants, human msh2 mutant cell lines (e.g., LoVo, HEC59, 2774) have a strong general mutator phenotype and strong microsatellite instability including instability at dinucleotide and mononucle- otide repeat loci (Bhattacharyya et al. 1994; Shibata etal. 1994; Boyer et al. 1995; Liu et al. 1995). Similar to the mutator phenotype of S. cerevisiae msh6 mutants, gtbp (msh6) mutant tumor lines (DLDl/HCTlS, VAC0543) have a strong mutator phenotype when the accumula- tion of mutations at HPRT (forward mutation assay) is measured, but their microsatellite instability phenotype is reduced compared to msh2 mutant tumor lines (Bhat- Figure 3. Model for mismatch recognition in S. cerevisiae. The various postulated complexes between MSH2 and either MSH3 tacharyya et al. 1994; Shibata et al. 1994; Eshleman et al. or MSH6 are illustrated interacting with either a single-base 1995; Papadopoulos et al. 1995). Little is known about substitution mispair or an insertioddeletion mispair; exactly the possible presence of msh3 mutations in either tumor which of the proteins in these complexesMSH2, MSH3, or cell lines or HNPCC families other than the fact that MSHGactually interacts with the mispaired base is not like GTBP (MSH6) mutations they must be rare because known. Also indicated is the previously described MLHI-PMS1 most HNPCC families studied have either msh2 or mlhl complex that interacts with the mispair recognition complex. mutations (Papadopoulos et al. 1995; Liu et al. 1996). The S. cerevisiae protein names are given as primary names; the The genetic properties of MSH2, MSH3, and MSH6 in human protein names are the same except for PMS1, which is S. cerevisiae have important implications for the analy- called PMS2 in humans, and MSH6, which has been called sis of mismatch repair-defective mutations and their as- GTBP or p160 in humans. (Reprinted, with permission, from Marsischky et al. 1996). sociation with cancer susceptibility. First, the redun- dancy of MSH3 and MSH6 compared to the apparently universal requirement for MSH2 in mismatch repair pro- vides an explanation for the high prevalence of msh2 mutation assays that detect a broader variety of muta- mutations in HNPCC families compared to mutations tion events, and in these experiments, the mutation rate in either GTBP (MSH6) or MSH3 (Papadopoulos et al. and mutation spectrum in the msh3 msh6 double mu- 1995, Liu et al. 1996). This is because independent mu- tant were essentially the same as that seen in the msh2 tations in both MSH3 and GTBP (MSH6) would be re- single mutant (Marsischky et al. 1996). These and other quired to produce the same mismatch repair defect as genetic data indicate that MSH3 and MSH6 encode re- that caused by mutations in MSH2. Second, selection for dundant activities that act in MSH2-dependent mis- tumors and syndromes (HNPCC) associated with high match repair (Johnson et al. 1996; Marsischky et al. degrees of microsatellite instability may preclude can- 19961. Consistent with this hypothesis, MSH2 forms cers and syndromes caused by msh6 or msh3 mutations. heterodimeric complexes with both MSH3 and MSH6 In contrast, one might expect to find msh3 or gtbp proteins (Marsischky et al. 1996). (msh6) mutations associated with cancers and cancer The analysis of S. cerevisiae MSH2, MSH3, and MSH6 susceptibility syndromes with mutator phenotypes that has led to the proposal of a model in which there are two more closely resemble the mutator phenotypes caused different pathways of MSH2-dependent mismatch repair: by msh3 and msh6 mutations in S. cerevisiae. Third, it is repair that is primarily specific for single-base substitu- possible that the mutator phenotype caused by msh3 or tion mispairs and requires a MSH2-MSH6 complex, and gtbp (msh6) mutations compared to msh2 mutations is repair that is primarily specific for insertion/deletion insufficient to cause cancer initiation or progression or mispairs and requires either a MSH2-MSH3 complex or that such mutations have lower penetrance than msh2 a MSH2-MSH6 complex (Fig. 3; Marsischky et al. 1996). mutations. Fourth, it is possible that the lack of another The sum of the mutagenesis data reported suggests that entirely MSH2-dependent process, rather than the loss of repair of single-base substitution mispairs requires the mismatch repair, is the underlying cause of cancer sus- MSH2-MSH6 complex, that repair of single-base inser- ceptibility. Fifth, it is possible that there are tumors and tionldeletion mispairs can utilize the MSH2-MSH6 or tumor cell lines showing microsatellite instability that 1436 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair are msh3, gtbp (msh6) double mutants. A particularly karyotic mismatch repair seen in regard to the MSH pro- intriguing possibility is that mutations in MSH3 and teins. GTBP (MSH6) segregate in the human population and are In both S. cerevisiae and human cells, there appear to relatively silent until msh3 gtbp (msh6) double mutant be at least two MutL homologs that are required for mis- individuals with increased cancer susceptibility are gen- match repair: MLHl and PMSl in S. cerevisiae, and erated. It would require two independent events to inac- MLHl and PMS2 (the closest homolog of S. cerevisiae tivate the wild-type MSH3 and GTBP (MSH6) alleles as PMS1) in human cells (Prolla et al. 1994; Kolodner 1995; expected in tumor cells from HNPCC individuals (Leach Li and Modrich 1995; Modrich and Lahue 1996). These et al. 1993; Hemminki et al. 1994; Borresen et al. 1995) proteins appear to exist in a 1:l complex (Prolla et al. and the occurrence of two such events might be infre- 1994bj Li and Modrich 1995). The purified human quent. However, the loss of the second allele of one of MLH1-PMS2 complex complements the in vitro mis- the genes, MSH3 or GTBP (MSHb), would cause a weak match repair defect caused by mlhl mutations found in mutator phenotype that could speed inactivation of the human tumor cell lines, indicating that it is functional remaining functional gene. It is also possible that MSH3I in mismatch repair (Li and Modrich 1995). In S. cerevi- msh3, GTBP (MSH6)lgtbp (msh6) cells might have an siae, this MLH1-PMS1 complex interacts with MSH2 increased mutation rate resulting from unlinked partial bound to a mispaired base, consistent with the idea that noncomplementation, which would increase the loss of the MLH1-PMS1 (human PMS2) complex plays the the second alleles of these genes. same role in mismatch repair in eukaryotes that MutL plays in bacteria. The available biochemical and genetic data (discussed in more detail below) and the various Homologs of the bacterial MutL proteins protein sequence homology relationships support the idea that the S. cerevisiae MLH1-PMS1 and human A variety of studies have documented a large number of MLH1-PMS2 complexes play equivalent roles in mis- eukaryotic genes encoding proteins that show a high de- match repair. gree of amino acid similarity with the bacterial MutL In human cells there is a third MutL homolog, PMS1, proteins (Figs. 3/41. It should be noted that the complete that has been implicated in mismatch repair by virtue of gene sequence is not known in all cases. Historically, the observation of a germ-line mutation in PMSl in a these proteins have been designated as either PMS or single patient with a family history of colon cancer MLH proteins because the first eukaryotic gene encoding (Nicolaides et al. 1994). The tumor from this patient was a MutL homolog was named PMSl before it had been not examined for either loss of the second PMS1 allele or cloned. These eukaryotic proteins, however, are all re- for microsatellite instability. Therefore, it is not clear lated to the bacterial MutL proteins. The available evi- whether PMS1 behaves like other mismatch repair and dence suggests that at least two of these proteins, MLHl tumor suppressor genes during tumor development or and PMSl (PMS2 in humans), are involved in mismatch whether pmsl mutations cause a mismatch repair defect repair. The involvement of two such proteins in mis- in tumors. At present, little is known about the bio- match repair, as compared to the involvement of a single chemical role of hPMSl in mismatch repair. Additional MutL protein in bacteria, parallels the complexity of eu- studies of human PMSl are clearly needed. The genetics of S. cerevisiae PMSl and MLHl and hu- man MLHl and PMS2 parallel each other to a great ex- tent. The available data indicate that mutations in S. PMS3 Human PMSR2 Human cerevisiae MLHl and PMSl cause a similar, strong gen- PMS4 Human eral mutator phenotype that is virtually the same as that PMSB Human caused by mutations in MSH2; no differences in the mu- PMSS Human PMS2 Human tator phenotypes caused by pmsl and mlhl mutations PMSR3 Human have yet been published (Williamson et al. 1985; Bishop PMS2 Mouse PMS6 Human et al. 1987; B. Kramer et al. 1989; Strand et al. 1993; PMSR6 Human Prolla et al. 1994a; Jeyaprakash et al. 1996). Interestingly, PMSl S.cerevisiae ' PMSl Human in a recent study on the effects of msh2 msh3, and pmsl (human pms2) mutations on crossing over between di- MUTL E.coli MUTL S.typhimuria vergent DNA sequences, PMSl (human PMS2) appeared MUTL H.influenza to play significantly less of a role in suppressing homol- HEXB S.pneumonia MUTL B.subtilis ogous recombination than MSH2 (Datta et al. 1996; Hunter et al. 1996). This suggests that defects in PMS1 MLHl S.cerevisiae I lMLH1 Human I MUTL V.cholera (human PMS2) might not be equivalent to defects in other mismatch repair genes (unfortunately MLHI was Figure 4. Family tree of known MutL homolog proteins. All of not evaluated in these studies; Datta et al. 1996; Hunter the sequences used for this construction were retrieved from et al. 1996). Analysis of human tumor cell lines with GenBank except S. cerevisiae MLH2, which was provided by mutations in either MLHI or PMS2 support this view: Tom Prolla and R. Michael Liskay (Prolla 1994). Some of the sequences used (hPMS5, hPMS6, hPMSR6, hPMS8) are not Such cell lines have a strong, global microsatellite insta- complete coding sequences. bility phenotype, and extracts prepared from such cell GENES & DEVELOPMENT 1437 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner lines are mismatch repair defective (Parsons et al. 1993; progress in the identification of exonucleases that might Bhattacharyya et al. 1994; Shibata et al. 1994; Boyer et al. play a role in eukaryotic mismatch repair. 1995; Li and Modrich 1995; Risinger et al. 1995). Fur- A 5' + 3' endo/exonuclease, S. cerevisiae RTH 1 thermore, the mismatch repair defect caused by mlhl (RAD27, YKL5 10)) has been proposed to function in mis- mutations can be complemented in vitro by the addition match repair (Johnson et al. 1995). Mutations in RTHl of the MLH1-PMS2 complex (Li and Modrich 1995). In (RAD27) cause a mutator phenotype (Johnson et al. 1995; addition, cells derived from homozygous pms2 mutant Reagan et al. 1995; Vallen and Cross 1995) that in one mice show microsatellite instability, consistent with a study was shown to be as strong as that caused by mu- mismatch repair defect (Baker et al. 1995). One striking tations in MSH2, MLH1, and PMS1 (Johnson et al. 1995). feature about the genetics of the human MLHl and This result seems at odds with mechanistic studies of PMS2 genes is that there is a strikingly lower frequency mismatch repair promoted by human cell extracts indi- of germ-line pms2 mutations compared to germ-line cating that, like the E. coli reaction, eukaryotic mis- mlhl mutations in HNPCC families, especially given match repair is likely to involve redundant 5' + 3' and that mutations in both genes appear to cause the same 3' + 5' exonucleases (Modrich 1991; Fang and Modrich type of mismatch repair defect (Boyer et al. 1995; Ris- 1993; Modrich and Lahue 1996). Double mutant combi- inger et al. 1995; Liu et al. 1996). These data may reflect nations of rthl and either msh2 mlhl, or pmsl showed trivial possibilities such as founder effects (Nystrom- a three- to fivefold synergistic effect (Johnson et al. 1995). Lahti et al. 1995), or DNA sequence contexts may make One possible interpretation of this result is that RTHl MLHl more mutable than PMS2. Alternately, these data functions in an entirely different pathway from MSH2, raise the intriguing possibilities that MLHl and PMS2 do MLH1, and PMSI . Additional genetic analysis of R THl not play equivalent roles in the cell even though they are (RAD27) has indicated that RTHl (RAD27) plays a role subunits of the same complex (see above discussion of in repair of other types of DNA damage, that RTHl Datta et al. 1996; Hunter et al. 1996), that pms2 muta- (RAD27) is a member of the RAD6 epistasis group that tions confer a selective disadvantage to the population functions in DNA damage tolerance, and that rthl compared to mlhl mutations (Baker et al. 1995), or that (rad27) mutants also have a cell cycle defect (Reagan et like S. cerevisiae MSH3 and MSH6, there is a function al. 1995; Vallen and Cross 1995). Studies of the S. cere- that is redundant with PMS2. The resolution of this visiae (called RTH1, RAD27, YKL510) and human question could provide important information about (called FEN-1, MF-1, exonuclease IV) protein have sug- how mismatch repair defects give rise to cancer suscep- gested that RTHl functions in the processing of 5' ends tibility. of Okazaki fragments and processing branched DNA structures formed by different DNA repair pathways (Ishimi et al. 1988; Harrington and Lieber 1994; Waga Other components required for mismatch repair and Stillman 1994; Waga et al. 1994; Hiraoka et al. 1995; It seems clear that HNPCC families, sporadic tumors, Sommers et al. 1995). There are several possible inter- and tumor cell lines that are associated with mismatch pretations of the data on RTHl with regard to a possible repair defects or that exhibit microsatellite instability role in mismatch repair. (1) RTHl functions as an exo- exist where it has not yet been possible to demonstrate nuclease during mismatch repair; however, the strong defects in known mismatch repair genes (da Costa et al. mutator phenotype of rthl mutants seems inconsistent 1995; Katabuchi et al. 1995; Liu et al. 1995, 1996; Pap- with the mechanistic studies implicating bidirectional adopoulos et al. 1995; Wijnen et al. 1995, 1996). This excision during mismatch repair. This suggests that if observation has provided additional interest in the iden- RTHl functions in mismatch repair, it likely functions tification of other components of eukaryotic mismatch in other repair reactions that act to reduce mutation repair. At present, there is little definitive additional in- rates. (2) RTHl might not function as an exonuclease but formation about the other proteins that function in eu- rather as an endonuclease that cleaves branched struc- karyotic MutHLS-like mismatch repair pathways. On tures that must be processed during different DNA repair the basis of our current understanding of the enzymes pathways, possibly including mismatch repair. (3) The that promote MutHLS mismatch repair in E. coli, obvi- mutator phenotype seen in rthl mutants is not attribut- ous candidates for proteins required for eukaryotic mis- able to a mismatch repair defect but is rather the result match repair are exonucleases, helicases, and enzymes of defects in the fidelity of DNA synthesis, or other re- required for DNA synthesis (Modrich 199 1 Kolodner pair defects. This latter possibility could account for the 1995; Modrich and Lahue 1996). A particularly interest- modest synergistic effect of rthl mutations in combina- ing unresolved issue is the identification of the proteins tion with msh2, mlhl, and pmsl mutations. Clearly cau- and signals that function in the recognition of the newly tion must be taken in the interpretation of the meaning replicated DNA strand during mismatch repair; this is of mutator phenotypes. because DNA methylation and MutH homologs seem A second candidate for an exonuclease involved in unlikely to play this role as they do in E. coli (Modrich mismatch repair was identified in a two-hybrid screen 1991; Kolodner 1995; Modrich and Lahue 1996). with S. cerevisiae MSH2 as a bait protein (D.X. Tishkoff, Whereas most of the progress reported in eukaryotic mis- A. Beddow, M.F. Kane, and R. Kolodner, in prep.). The S. match repair enzymology has been in the area of the cerevisiae EX01 gene was identified as encoding a pro- MSH and MLH proteins, there has been some recent tein that interacts with both S. cerevisiae and human 1438 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair MSH2, and a S. cerevisiae MSH2-EX01 complex was 1995). One of several possible explanations for these data identified in vivo by coimmunoprecipitation. S. cerevi- is that there could be a separate, less-efficient repair siae EX01 appears to be a homolog of the Schizosaccha- pathway that selectively repairs insertionldeletion mi- romyces pom be 5' -+ 3' exonuclease EX0 1 (Szankasi spairs by deletion. Consistent with this view, analysis of and Smith 1992, 1995; D.X. Tishkoff, A. Beddow, M.F. mismatch repair catalyzed in extracts of human tumor Kane, and R. Kolodner, in prep.), and characterization of cell lines has suggested there may be an MLHI -indepen- the overproduced and purified S. cerevisiae EXOl has dent mismatch repair pathway that promotes the repair indicated that it preferentially degrades double-stranded of some insertionldeletion mispairs >3 nucleotides in DNA compared with single-stranded DNA. Mutations length and that repair of some such substrates may also in S. cerevisiae EX01 cause a weak general mutator phe- be MSH2-independent (Umar et al. 1994a). notype and low levels of microsatellite instability, con- More recently, several workers have detected S. cere- sistent with the redundant exonuclease proposal. Fur- visiae and human mispair binding activities present in S. thermore, analysis of double mutants was consistent cerevisiae msh2, msh3, and msh4 mutants and human with EX01 functioning in a MSH2-dependent pathway msh2 and gtbp (msh6) mutants (Miret et al. 1996; (D.X. Tishkoff, A. Beddow, M.F. Kane, and R. Kolodner, O1Regan et al. 1996). Other studies have detected activ- in prep.). S. pombe ex01 mutants also have a mutator ities that nick DNA at the site of a mispair (Chang and phenotype, and S. pombe EXOl has been proposed to Lu 1991; Yao and Kom 1994; Yeh et al. 1994). However, function in either a general MutHLS-like mismatch re- it has not been established whether these activities play pair system or a more specific mismatch repair system a role in a general mismatch repair pathway such as the like the base-specific systems described in bacteria MutHLS-like pathway or whether these activities act in (Szankasi and Smith 1995). At this stage of analysis ad- other pathways such as crossing over during meiotic re- ditional studies are clearly needed to define the potential combination, which is known to require MSH4 and roles of RTHl (RAD27) and EXOl in mismatch repair MSH5 (Ross-Macdonald and Roeder 1994; Holling- and to identify additional exonucleases that might be sworth et al. 1995), or in base-specific mismatch repair required. (Modrich 199 1 ; Kolodner 1995; Modrich and Lahue 1996). Whether these types of activities play roles in other mismatch repair pathways and whether alternate Is there another eukaryotic general mismatch repair mismatch repair pathways that recognize a broad spec- pathway? trum of mispaired bases including larger insertionldele- A number of lines of investigation have suggested that tion mispairs exist remain important unanswered ques- there may be another eukaryotic mismatch repair path- tions at this time. way that has an overlapping specificity with the MutHLS-like pathway. When gene conversion was ana- lyzed in S. cerevisiae pmsl, msh2, or mlhl mutants by Conclusions use of tetrad analysis, it was observed that there was a A considerable amount of progress has been made in the low, residual level of gene conversion independent of the identification of components of eukaryotic MutHLS-like marker analyzed (single-base substitution mutations or mismatch repair pathways. These pathways, however, small insertionldeletion mutations) (Williamson et al. appear more complex than their bacterial counterparts. 1985; Reenan and Kolodner 1992a; Alani et al. 1994; Furthermore, not all of the proteins that function in Prolla et al. 1994a). This could occur for two possible these eukaryotic pathways have been identified, nor has reasons: Either a proportion of gene conversion events it been resolved how many such repair pathways might are caused by gap repair and do not involve the formation exist. Particular challenges for the future include the of a heteroduplex recombination intermediate (discussed complete identification of the components of these re- in Alani et al. 1994), or there is an inefficient mismatch pair pathways, the understanding of how defects in these repair pathway functional in pmsl, msh2, or mlhl null pathways lead to cancer susceptibility, and the unravel- mutants. Similarly, when transformation assays were ing of the genetics of these pathways in model organ- used to analyze mismatch repair directly, not all appar- isms, which may provide greater insights into the role of ent mismatch repair was eliminated by mutations in inherited and acquired repair defects in cancer suscepti- mismatch repair genes (Bishop et al. 1987, 1989; B. bility in the human population. Kramer et al. 1989). Furthermore, the frequency of inser- tion events approximately equaled the frequency of de- letion events when insertionldeletion mispairs were re- paired after transformation of wild-type S. cerevisiae Acknowledgments cells, whereas there was a strong disparity in favor of I thank Richard Boland, Tom Kunkel, Gerry Marsischky, Dan deletion events among the fraction of insertionldeletion Tishkoff, Geoff Wahl, and Jean Wang for suggestions and com- mispairs that appeared to be repaired in a pmsl mutant ments on the manuscript. I particularly acknowledge the hos- (Bishop and Kolodner 1986; Bishop et al. 1987). A similar pitality of the Salk Institute for Biological Studies during the shift toward deletion events in mismatch repair mutants preparation of this paper. Work in my laboratory on mismatch was seen when alterations in GT tract length resulting repair has been supported by National Institutes of Health from replication errors were analyzed (Strand et al. 1993, grants GM50006 and CA6715 1. GENES & DEVELOPMENT 1439 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner References Eshleman, J.R., E.Z. Lang, G.K. Bowerfind, R. Parsons, B. Vo- gelstein, J.K.V. Wilson, M.L. Veigl, W.D. Sedwick, and S.D. Alani, E., R.A.G. Reenan, and R. Kolodner. 1994. Mismatch Markowitz. 1995. Increased mutation rate at the hprt locus repair proteins directly affect gene conversion in Saccharo- accompanies microsatellite instability in colon cancer. On- myces cerevisiae by regulating heteroduplex tract length. cogene 10: 33-37. Genetics 137: 19-39. Fang, W.-H. and P. Modrich. 1993. Human strand-specific mis- Alani, E., N.-W. Chi, and R.D. Kolodner. 1995. The Saccharo- match repair occurs by a bidirectional mechanism similar to myces cerevisiae MSH2 protein specifically binds to duplex that of the bacterial reaction. \. Biol. Chem. 268: 1 1838-1 1844. oligonucleotides containing mismatched DNA base pairs Feng, W.Y., E.H. Lee, and J.B. Hays. 199 1. Recombinogenic pro- and loop insertions. Genes & Dev. 9: 234-247. cessing of UV-light photoproducts in nonreplicating phage Anthoney, D.A., A.J. McIlwrath, W.M. Gallagher, A.R.M. Edlin, DNA by the Escherichia coli methyl-directed mismatch re- and R. Brown. 1996. Microsatellite instability, apoptosis, pair system. Genetics 129: 1007-1020. and loss of p53 function in drug-resistant tumor cells. Can- Fishel, R. and R.D. Kolodner. 1995. Identification of mismatch cer Res. 56: 1374-1381. repair genes and their role in the development of cancer. Baker, S.M., C.E. Bronner, I. Zhang, A. Plug, M. Robatzek, G. Curr. Opin. Genet. Dev. 5: 382-395. Warren, E.A. Elliott, J. Yu, T. Ashley, N. Arnheim, R.A. Fla- Fishel, R.A., M.K. Lescoe, M.R.S. Rao, N. Copland, N. Jenkins, vell, and R.M. Liskay. 1995. Male mice defective in the DNA J. Garber, M. Kane, and R. Kolodner. 1993. The human mu- mismatch repair gene PMS2 exhibit abnormal chromosome tator gene homolog MSH2 and its association with heredi- synapsis in meiosis. Cell 82: 309-320. tary nonpolyposis colon cancer. Cell 75: 1027-1038. Bhattacharyya, N.P., A. Skandalis, A. Ganesh, J. Groden, and M. Fishel, R.A., A. Ewel, S. Lee, M.K. Lescoe, and J. Griffith. 1994. Meuth. 1994. Mutator phenotype in human colorectal car- Binding of mismatched microsatellite DNA sequences by cinoma cell lines. Proc. Natl. Acad. Sci. 91: 63 19-6323. the human MSH2 protein. Science 266: 1403-1405. Bishop, D.K. and R.D. Kolodner. 1986. Repair of heteroduplex Fram, R. J., P.S. Cusick, J.M. Wilson, and M.G. Marinus. 1985. plasmid DNA after transformation into Saccharomyces cer- Mismatch repair of cis-Diamminedichloroplatinum(II)-In- evisiae. Mol. Cell. Biol. 6: 3401-3409. duced DNA damage. Mol. Pharmacol. 28: 51-55. Bishop, D.K., M.S. Williamson, S. Fogel, and R.D. Kolodner. Friedberg, E.C., G.C. Walker, and W. Siede. 1995. DNA repair 1987. The role of heteroduplex correction in gene conversion and mutagenesis. ASM Press, Washington, D.C. in Saccharomyces cerevisiae. Nature 328: 362-364. Grilley, M., J. Griffith, and P. Modrich. 1993. Bidirectional ex- Bishop, D.K., J. Andersen, and R.D. Kolodner. 1989. Specificity cision in methyl-directed mismatch repair. [. Biol. Chem. of mismatch repair following transformation of Saccharo- 268: 11830-1 1837. myces cerevisiae with heteroduplex plasmid DNA. Proc. Harrington, J. J. and M.R. Lieber. 1994. Functional domains Natl. Acad. Sci. 86: 3713-3717. within FEN-1 and RAD2 define a family of structure-specific Borresen, A.-L., R.A. Lothe, G.I. Meling, S. Lystad, P. Morrison, endonucleases: Implications for nucleotide excision repair. J. Lipford, M.F. Kane, T.O. Rognum, and R.D. Kolodner. Genes & Dev. 8: 1344-1355. 1995. Somatic mutations in the hMSH2 gene in microsatel- Hawn, M.T., A. Umar, J.M. Caretthers, G. Marra, T.A. Kunkel, lite unstable colorectal carcinomas. Hum. Mol. Genet. R.C. Boland, and M. Koi. 1995. Evidence for a connection 11: 2065-2072. between the mismatch repair system and the G2 cell cycle Boyer, J.C., A. Umar, J. Risinger, M.F. Kane, J. Lipford, J.C. Bar- checkpoint. Cancer Res. 55: 3721-3725. rett, R.D. Kolodner, and T.A. Kunkel. 1995. Microsatellite Hemminki, A., P. Peltomaki, J.-P. Mecklin, H. Jarvinen, R. Sa- instability, mismatch repair deficiency and genetic defects lovaara, M. Nystrom-Lahti, A. de la Chapelle, and L.A. Aal- in human cancer cell lines. Cancer Res. 55: 6062-6070. tonen. 1994. Loss of the wild-type MLHl gene is a feature of Bronner, C.E., S.M. Baker, P.T. Morrison, G. Warren, L.G. hereditary nonpolyposis colorectal cancer. Nature Genet. Smith, M.K. Lescoe, M. Kane, C. Earabino, J. Lipford, A. 8: 405-41 1. Lindblom, P. Tannergard, R.J. Bollag, A.R. Godwin, D.C. Hiraoka, L.R., J.J. Harrington, D.S. Gerhard, M.R. Lieber, and Ward, M. Nordenskjold, R. Fishel, R. Kolodner, and R.M. C.-L. Hsieh. 1995. Sequence of human FEN-1, a structure- Liskay. 1994. Mutation in the DNA mismatch repair gene specific endonuclease, and chromosomal location of the homologue hMLHl is associated with hereditary nonpoly- gene (FENI) in mouse and human. Genomics 25: 220-225. posis colon cancer. Nature 368: 258-261. Hollingsworth, N.M., L. Ponte, and C. Halsey. 1995. MSHS, a Chang, D.Y. and A.-L. Lu. 1991. Base mismatch-specific endo- novel MutS homolog, facilitates meiotic reciprocal recom- nuclease activity in extracts from Saccharomyces cerevi- bination between homologs in Saccharomyces cerevisiae siae. Nucleic Acids Res. 19: 4761-4766. but not mismatch repair. Genes & Dev. 9: 1728-1739. da Costa, L.T., B. Liu, W. el-Deiry, S.R. Hamilton, K.W. Kinzler, Hunter, N., S.R. Chambers, E.J. Louis, and R.H. Borts. 1996. The B. Vogelstein, S. Markowitz, J.K. Willson, A. de la Chapelle, mismatch repair system contributes to meiotic sterility in K.M. Downey, and A.G. So. 1995. Polymerase delta variants an interspecific yeast hybrid. EMBO \. 15: 1726-1 733. in RER colorectal tumours [letter]. Nature Genet. 9: 1&11. Ishimi, Y., A. Claude, P. Bullock, and J. Hurwitz. 1988. Com- Datta, A., A. Adjiri, L. New, G.F. Crouse, and S. Jinks-Robert- plete enzymatic synthesis of DNA containing the SV40 or- son. 1996. Crossovers between diverged sequences are regu- igin of replication. \. Biol. Chem. 263: 19723-19733. lated by mismatch repair proteins in yeast. Mol. Cell. Biol. Jeyaprakash, A., R. DasGupta, and R. Kolodner. 1996. Saccha- 16: 1085-1093. romyces cerevisiae pms2 mutations are alleles of MLHl and Drummond, J.T., G.-M. Li, M.J. Longley, and P. Modrich. 1995. pms2-2 corresponds to an HNPCC causing missense muta- Mismatch recognition by an hMSH2-GTBP heterodimer tion. Mol. Cell Biol. (in press). and differential repair defects in tumor cells. Science Johnson, R.E., K.K. Gopala, L. Prakash, and S. Prakash. 1995. 268: 1909-1912. Requirement of the yeast RTHl 5' to 3' exonuclease for the Eshleman, J.R. and S.D. Markowitz. 1995. Microsatellite insta- stability of simple repetitive DNA. Science 269: 238-240. bility in inherited and sporadic neoplasms. Cur. Opin. On- Johnson, R.E., G.K. Kovvali, L. Prakash, and S. Prakash. 1996. CO~. 7: 83-89. Requirement of the yeast MSH3 and MSH6 genes for MSH2- 1440 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair dependent genomic stability. J. Biol. Chem. 271: 7285-7288. Rev. Biochem. 65: 101-133. Karran, P. and M.G. Marinus. 1982. Mismatch correction of New, L., K. Liu, and G.F. Crouse. 1993. The yeast gene MSH3 06-methylguanine residues in E. coli DNA. Nature defines a new class of eukaryotic MutS homologues. Mol. & 296: 868-869. Gen. Genet. 239: 97-108. Katabuchi, H., B. van Rees, A.R. Lambers, B.M. Ronnett, M.S. Nicolaides, N.C., N. Papadopoulos, B. Liu, Y. Wei, K.C. Carter, Blazes, F.S. Leach, K.R. Cho, and L. Hedrick. 1995. Muta- S.M. Ruben, C.A. Rosen, W.A. Haseltine, R.D. Fleischmann, tions in DNA mismatch repair genes are not responsible for C.M. Fraser, M.D. Adams, J.C. Venter, M.G. Dunlop, S.R. microsatellite instability in most sporadic endometrial car- Hamilton, G.M. Petersen, A. de la Chapelle, B. Vogelstein, cinomas. Cancer Res. 55: 5556-5560. and K. Kinzler. 1994. Mutations of two PMS homologues in Kolodner, R.D. 1995. Mismatch repair: Mechanisms and rela- hereditary nonpolyposis colon cancer. Nature 371: 75-80. tionshlp to cancer susceptibility. Trends Biochem. 20: 397402. Nystrom-Lahti, M., P. Kristo, N.C. Nicoladies, S.-Y. Chang, Kramer, B., W. Kramer, M.S. Williamson, and S. Fogel. 1989. L.A. Aaltonen, A.-L. Moisio, H.J. Jarvinen, J.-P. Mecklin, K. Heteroduplex DNA correction in Saccharomyces cerevisiae Kinzler, B. Vogelstein, A. de la Chapelle, and P. Peltomaki. is mismatch specific and requires functional PMS genes. 1995. Founding mutations and Alu-mediated recombination Mol. Cell. Biol. 9: 4432-4440. in hereditary colon cancer. Nature Med. 1: 6054-6058. Kramer, W., B. Kramer, M.S. Williamson, and S. Fogel. 1989. OIRegan, N.E., P. Branch, P. Macpherson, and P. Karran. 1996. Cloning and nucleotide sequence of DNA mismatch repair hMSH2-independent DNA mismatch recognition by human gene PMSl from Saccharomyces cerevisiae: Homology of proteins. 1. Biol. Chem. 271: 1789-1796. PMSl to procaryotic MutL and HexB. 1. Bacterial. Palombo, F., P. Gallinari, I. Iaccarino, T. Lettieri, M. Hughes, A. 171: 5339-5346. D'Arrigo, 0. Truong, J.J. Hsuan, and J. Jiricny. 1995. GTBP, Lahue, R.S., K.G. Au, and P. Modrich. 1989. DNA mismatch a 160 kD protein essential for mismatch binding activity in correction in a defined system. Science 245: 160-164. human cells. Science 268: 1912-1914. Leach, F.S., N.C. Nicolaides, N. Papadopoulos, B. Liu, J. Jen, R. Papadopoulos, N., N.C. Nicolaides, Y.-F. Wei, S.M. Ruben, K.C. P. Peltomaki, P. Sistonen, L.A. Aaltonen, M. Nys- Parsons, Carter, C.A. Rosen, W.A. Haseltine, R.D. Fleischmann, C.M. trom-Lahti, X.-Y. Guan, J. Zhang, P.S Meltzer, J.-W. Yu, F.-T. Fraser, M.D. Adams, J.C. Venter, S.R. Hamilton, G.M. Pe- Kao, D.J. Chen, K.M. Cerosaletti, R.E.K. Fournier, S. Todd, tersen, P. Watson, H.T. Lynch, P. Peltomaki, J.-P. Mecklin, T. Lewis, R.J. Leach, S.L. Naylor, J. Weissenbach, J.-P. Meck- A. de la Chapelle, K.W. Kinzler, and B. Vogelstein. 1994. lin, H. Jarvinen, G.M. Petersen, S.R. Hamilton, J. Green, J. Mutation of a mutL homolog in hereditary colon cancer. Jass, P. Watson, H.T. Lynch, J.M. Trent, A. de la Chapelle, Science 263: 1625-1629. K.W. Kinsler, and B. Vogelstein. 1993. Mutations of a mutS Papadopoulos, N., N.C. Nicolaides, B. Liu, R.E. Parsons, C. Len- homolog in hereditary nonpolyposis colorectal cancer. Cell gauer, F. Palombo, A. D'Arrigo, S. Markowitz, J.K.V. Wilson, 75: 1215-1225. K.W. Kinzler, J. Jiricny, and B. Vogelstein. 1995. Mutations Li, G.-M. and P. Modrich. 1995. Restoration of mismatch repair of GTBP in genetically unstable tumors. Science 268: 1915- to nuclear extracts of H6 colorectal tumor cells by a hetero- 1917. dimer of human MutL homologs. Proc. Natl. Acad. Sci. Parsons, R., G.-M. Li, M.J. Longley, W.-H. Fang, N. Papadopou- 92: 1950-1954. los, J. Jen, A. de la Chapelle, K.W. Kinzler, B. Vogelstein, and Liu, B., M.C. Nicolaides, S. Markowitz, J.K.V. Willson, J.R. Par- P. Modrich. 1993. Hypermutability and mismatch repair de- sons, J. Jen, N. Papadopolous, P. Peltomaki, A. de la ficiency in RER + tumor cells. Cell 75: 1227-1236. Chapelle, S.R. Hamilton, K.W. Kinzler, and B. Vogelstein. Proffitt, J.H., J.R. Davie, D. Swinton, and S. Hattman. 1984. 1995. Mismatch repair gene defects in sporadic colorectal 5-Methylcytosine is not detectable in Saccharomyces cere- cancers with microsatellite instability. Nature Genet. 9: 48- Mol. Cell. Biol. 4: 985-988. visiae. Prolla, T.A. 1994. Dual requirement in yeast DNA mismatch Liu, B., R. Parsons, N. Papadopoulos, N.C. Nicolaides, H.T. repair for MLHl and PMS1, two homologs of bacterial MutL. Lynch, P. Watson, J.R. Jass, M. Dunlop, A. Wyllie, P. Pelto- Ph.D. thesis. Yale University, New Haven, CT. maki, A. de la Chapelle, S.R. Hamilton, R. Vogelstein, and Prolla, T.A., D.-M. Christie, and R.M. Liskay. 1994a. A require- K.W. Kinzler. 1996. Analysis of mismatch repair genes in ment in yeast DNA mismatch repair for MLHl and PMSI, hereditary non-polyposis colorectal cancer patients. Nature two homologs of the bacterial mutL gene. Mol. Cell. Biol. Med. 2: 169-174. 14: 407-415. Marra, G. and C.R. Boland. 1995. Hereditary nonpolyposis co- Prolla, T.A., Q. Pang, E. Alani, R.D. Kolodner, and R.M. Liskay. lorectal cancer: The syndrome, the genes and historical per- 199413. Interactions between the MSH2, MLHl and PMSl spectives. f. Natl. Cancer Inst. 87: 11 14-1 125. proteins during the initiation of DNA mismatch repair. Sci- Marsischky, G.T., N. Filosi, M.F. Kane, and R. Kolodner. 1996. ence 265: 1091-1093. Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 Rayssiguier, C., D.S. Thaler, and M. Radman. 1989. The barrier in MSH2-dependent mismatch repair. Genes & Dev. to recombination between Escherichia coli and Salmonella 10: 407-420. typhimurium is disrupted in mismatch-repair mutants. Na- Mellon, I. and G.N. Champe. 1996. Products of DNA mismatch ture 342: 396-40 1. repair genes mutS and mutL are required for transcription- Reagan, M.S., C. Pittenger, W. Siede, and E.C. Friedberg. 1995. coupled nucleotide excision repair of the lactose operon in Characterization of a mutant strain of Saccharomyces cere- Escherichia coli. Proc. Natl. Acad. Sci. 93: 1292-1297. visiae with a deletion of the RAD27 gene, a structural ho- Miret, J.J., B.O. Parker, and R.S. Lahue. 1996. Recognition of mologog of the RAD2 nucleotide excision repair gene. J. Bac- DNA insertionldeletion mismatches by an activity in Sac- teriol. 177: 364-371. charomyces cerevisiae. Nucleic Acids Res. 24: 721-729. Reenan, R.A.G. and R.D. Kolodner. 1992a. Characterization of Modrich, P. 1991. Mechanisms and biological effects of mis- insertion mutations in the Saccharomyces cerevisiae MSHl match repair. Annu. Rev. Genet. 25: 229-253. and MSH2 genes: Evidence for separate mitochondria1 and Modrich, P. and R. Lahue. 1996. Mismatch repair in replication nuclear functions. Genetics 132: 975-985. fidelity, genetic recombination and cancer biology. Annu. . 1992b. Isolation and characterization of two Saccharo- GENES & DEVELOPMENT 1441 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner myces cerevisiae genes encoding homologs of the bacterial L. Bisgaard, J. Mohr, R. Kolodner, and R. Fodde. 1996. Ma- HexA and MutS mismatch repair proteins. Genetics jority of hMLH1 mutations responsible for hereditary non- 132: 963-973. polyposis colorectal cancer (HNPCC) cluster at the exonic region 15-16. Am. 1. Hum. Genet. 58: 300-307. Risinger, J.I., A. Umar, J.C. Barrett, and T.A. Kunkel. 1995. A Williamson, M.S., J.C. Game, and S. Fogel. 1985. Meiotic gene hPMS2 mutant cell line is defective in strand-specific mis- conversion mutants in Saccharomyces cerevisiae. I. Isola- match repair. J. Biol. Chem. 270: 18183-18186. tion and characterization of pmsl -1 and pmsl -2. Genetics Ross-Macdonald, P. and G.S. Roeder. 1994. Mutation of a mei- osis-specific MutS homolog decreases crossing over but not 110: 609-646. mismatch correction. Cell 79: 1069-1080. Yao, M. and Y.W. Kom. 1994. Strand-specific cleavage of mis- match-containing DNA by deoxyinosine 3'-endonuclease Selva, E.M., L. New, G.F. Crouse, and R.S. Lahue. 1995. Mis- from Escherichia coli. 1. Biol. Chem. 269: 31390-31396. match correction acts as a barrier to homeologous recombi- nation in Saccharomyces cerevisiae. Genetics 139: 1175- Yeh, Y.C., H.F. Liu, C.A. Ellis, and A.-L. Lu. 1994. Mammalian 1188. topoisomerase I has base mismatch nicking activity. 1. Biol. Shibata, D., M.A. Peinado, S. Ionov, S. Malkhosyan, and M. Chem. 269: 15498-1 5504. Perucho. 1994. Genomic instability in repeated sequences in an early somatic event in colorectal tumorigenesis that per- sists after transformation. Nature Genet. 6: 273-28 1. Sommers, C.H., E.J. Miller, B. Dujon, S. Prakash, and L. Prakash. 1995. Conditional lethality of null mutations in RTHl that encodes the yeast counterpart of a mammalian 5'-to 3'-exo- nuclease required for lagging strand DNA synthesis in re- constituted systems. I. Biol. Chem. 270: 4193-4196. Strand, M., T.A. Prolla, R.M. Liskay, and T.D. Petes. 1993. De- stabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature 365: 274-276. 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. 92: 10418-10421. Szankasi, P. and G.R. Smith. 1992. A DNA exonuclease induced during meiosis of Schizosaccharomyces pombe. 1. Biol. Chem. 267: 3014-3023. . 1995. A role for exonuclease I from S. pombe in muta- tion avoidance and mismatch correction. Science 267: 1166- Umar, A., J.C. Boyer, and T.A. Kunkel. 1994a. DNA loop repair by human cell extracts. Science 266: 814-816. Umar, A., J.C. Boyer, D.C. Thomas, D.C. Nguyen, J.I. Risinger, J. Boyd, Y. Ionov, M. Perucho, and T.A. Kunkel. 1994b. De- fective mismatch repair in extracts of colorectal and endom- etrial cancer cell lines exhibiting microsatellite instability. 1. Biol. Chem. 269: 14367-14370. Vallen, E.A. and R.R. Cross. 1995. Mutations in RAD27 define a potential link between G1 cyclins and DNA replication. Mol. Cell Biol. 15: 4291-4302. Waga, S. and B. Stillman. 1994. Anatomy of a DNA replication fork revealed by reconstitution of SV40 DNA replication in vitro. Nature 369: 207-212. Waga, S., G. Bauer, and B. Stillman. 1994. Reconstitution of complete SV40 replication with purified replication factors. 1. Biol. Chem. 269: 10923-10934. Watanabe, A., M. Ikejima, N. Suzuki, and T. Shimada. 1996. Genomic organization and expression of the human MSH3 gene. Genomics 31: 311-318. Wijnen, J., H. Vassen, P. Meera Kahn, F.H. Menko, H. van der Klift, C. van Leeuwen, M. van den Broek, I. van Leeuwen- Comelisse, F. Nagengast, A. Meijers-Heijboer, D. Lindhout, G. Griffioen, A. Cats, J. Kleibeuker, L. Varesco, L. Bertario, M.L. Bisgaard, J. Mohr, and R. Fodde. 1995. Seven new mu- tations in hMSH2, an HNPCC gene, identified by denaturing gradient gel electrophoresis. Am. J. Hum. Genet. 56: 1060- Wijnen, J., P. Meera Kahn, H. van der Klift, H. Vasen, F.H. Menko, F. Nagengast, A. Meijers-Heijboer, D. Lindhout, G. Griffioen, A. Cats, J. Kleibeuker, L. Varesco, L. Bertario, M.- 1442 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press R Kolodner Genes Dev. 1996, 10: Access the most recent version at doi:10.1101/gad.10.12.1433 This article cites 84 articles, 49 of which can be accessed free at: References http://genesdev.cshlp.org/content/10/12/1433.full.html#ref-list-1 License Receive free email alerts when new articles cite this article - sign up in the box at the top Email Alerting right corner of the article or click here. Service Copyright © Cold Spring Harbor Laboratory Press http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Genes & Development Unpaywall

Biochemistry and genetics of eukaryotic mismatch repair.

Kolodner, R
Genes & Development – Jun 15, 1996
Download PDF
11 pages

Loading next page...
 
/lp/unpaywall/biochemistry-and-genetics-of-eukaryotic-mismatch-repair-gY0nXwaSdN

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0890-9369
DOI
10.1101/gad.10.12.1433
Publisher site
See Article on Publisher Site

Abstract

Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press REVIEW Biochemistry and genetics of eukaryotic mismatch repair Richard Kolodner Charles A. Dana Division of Human Cancer Genetics, Dana-Farber Cancer Institute, Boston, Massachusetts 021 15 USA and Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 021 15 USA The process of mismatch repair was first postulated to match repair, base-specific mismatch repair systems, explain the results of experiments on genetic recombi- and cancer genetics, see other recent reviews (Modrich nation and bacterial mutagenesis. Mismatch repair has 1991; Eshleman and Markowitz 1995; Fishel and Kolod- long been known to play a major role in two cellular ner 1995; Friedberg et al. 1995; Kolodner 1995; Marra processes: (1) the repair of errors made during DNA rep- and Boland 1995; Modrich and Lahue 1996). lication or as the result of some types of chemical dam- age to DNA and DNA precursors; and (2) the processing The E. coli MutHLS mismatch repair pathway of recombination intermediates to yield new configura- tions of genetic markers. More recent studies have sug- The E. coli MutHLS mismatch repair pathway is a gen- gested that mismatch repair may also be crucial for (1) eral DNA repair pathway that recognizes and repairs all the regulation of recombination events between diver- single-base mispairs except C.C (Modrich 199 1 Fishel gent DNA sequences that could result in different types and Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; of genetic instability (Rayssiguier et al. 1989; Selva et al. Modrich and Lahue 1996). It also repairs small insertion1 1995; Datta et al. 1996), (2) some types of nucleotide deletion mispairs, although it may not efficiently recog- excision repair responsible for repair of physicallchemi- insertionldeletion mispairs that have more nize most cal damage to DNA (Karran and Marinus 1982; Fram et than 4 unpaired bases. The basic repair reaction cata- al. 1985; Feng et al. 1991; Mellon and Champe 1996), and lyzed by this pathway is understood in considerable de- (3) participating in a cell-cycle checkpoint control sys- tail because it has been reconstituted in vitro with DNA tem by recognizing certain types of DNA damage and substrates containing mispaired bases, MutH, MutL, triggering cell-cycle arrest or other responses to DNA MutS, and UvrD (helicase 11) proteins, DNA polymerase damage (Hawn et al. 1995; Anthoney et al. 1996). I11 holoenzyme, DNA ligase, single-strand DNA-binding The most extensively characterized general mismatch protein (SSB), and any one of the single-stranded DNA repair system is the Escherichia coli MutHLS system, exonucleases-Exo I, Exo VII, and RecJ protein (Fig. 1; which repairs a broad spectrum of mispaired bases and Lahue et al. 1989; Modrich 1991; Grilley et al. 1993; has been reconstituted with purified enzymes. Eukary- Fishel and Kolodner 1995; Friedberg et al. 1995; Kolod- otes are known to contain a mismatch repair system that ner 1995; Modrich and Lahue 1996). The reaction in- has at least some components that are highly related to volves mismatch-dependent nicking of the unmethy- key components of the bacterial MutHLS mismatch re- lated strand at a hemimethylated GATC site and degra- pair system. The observation that defects in mismatch dation from the nick past the mismatch followed by repair genes are linked to both inherited cancer suscep- resynthesis. The roles of many of the proteins that func- tibility and some sporadic cancers has generated consid- tion in this reaction have been elucidated. The MutS erable interest in the gene products that function in eu- protein binds to DNA at the site of a mispaired base and karyotic mismatch repair. The goal of this review is to is responsible for mismatch recognition. No activity has discuss recent studies on the mechanisms of MutHLS- been assigned to MutL, although it interacts with MutS like mismatch repair in the yeast Saccharomyces cere- bound to a mispaired base and is required for activation visiae and in humans and to relate insights derived from of MutH. MutH is an endonuclease that nicks hemime- these studies to human cancer genetics. Given space thylated DNA on the unmethylated strand when acti- constraints, it is difficult to cover everything known vated by MutS and MutL in the presence of a mismatch. about mismatch repair or to reference all of the relevant The requirement for MutH in the reaction can be re- work that has been done in this area. However, a brief placed by a pre-existing nick in the DNA. The repair overview of the E. coli MutHLS pathway is presented reaction can utilize hemimethylated sites that are either below to allow comparison of the E. coli and eukaryotic 5' or 3' to the mispair. Excision requires UvrD (helicase mismatch repair pathways and proteins. For more de- 11) and one of the single-stranded DNA exonucleases- tailed information, particularly related to bacterial mis- Exo 1 (3' exonuclease), Exo VII (3' and 5' exonuclease), GENES & DEVELOPMENT 10:1433-1442 O 1996 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/96 $5.00 1433 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner vincing evidence has implicated DNA methylation in strand discrimination during mismatch repair in other eukaryotes (Modrich 1991; Fishel and Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; Modrich and Lahue ///A /- MutH 1996). Thus, it seems possible that the mechanism of strand discrimination in eukaryotes will prove to be quite different from that observed in E. coli. Exo I, Exo VII or RecJ, Helicase II, MutHLS-like mismatch repair in eukaryotes DNA Pol Ill, SSB and 3' DNA ligase Two lines of experimentation have indicated that eu- karyotes have a broad-spectrum mismatch repair system related to the bacterial MutHLS system. First, a series of genetic studies led to the identification of s.' cerevisiae homologs of the bacterial mutL and mutS gene products and the demonstration that these gene products were required for mismatch repair (Williamson et al. 1985; Figure 1. Illustration of the action of the E. coli MutHLS mis- Bishop et al. 1987; W. Kramer et al. 1989; Reenan and match repair system on a mispair at a replication fork. Repair is Kolodner 1992a; Reenan and Kolodner 1992b; Kolodner initiated by binding of MutS protein to a mismatch. The sub- 1995). Second, biochemical studies of higher eukaryotic sequent binding of MutL to MutS is required to activate MutH, which then nicks the unmethylated strand of DNA at hemim- cells have demonstrated in vitro nick-directed revair of a ethylated GATC sites. Nicking of the unmethylated strand is variety of different mispaired bases similar to the repair then followed by the excision from the nick to the mispair and reactions catalyzed by the E. coli MutHLS system (Fang resynthesis to fill in the resulting gap. These interactions result and Modrich 1993; Modrich and Lahue 1996). More re- in the coupling of mismatch repair to DNA replication, so that cently, biochemical studies have demonstrated that mismatches formed during DNA replication are repaired using these mismatch repair reactions are dependent on eu- the methylated parental strand as template, resulting in a re- karyotic homologs of the bacterial mutL and mutS gene duction of misincorporation errors. (Adapted from Kolodner products (Umar et al. 1994; Boyer et al. 1995; Drum- 1995.1 mond et al. 1995: Li and Modrich 19951. A considerable body of evidence indicates that this plays an important role in maintaining replication fidelity, pro- and RecJ protein (5' exonucleasej-depending on cessing recombination intermediates, and regulating re- whether the nicked, unmethylated site is 5' or 3' to the combination in response to sequence divergence. These mispair. Once excision has occurred, resynthesis is me- subjects, however, are beyond the scope of this review. diated by DNA polymerase I11 holoenzyme, SSB, and DNA ligase. Homologs of the bacterial MutS proteins A crucial feature of the MutHLS mismatch repair pathway is its ability to preferentially repair the daugh- A series of studies has demonstrated that there are at ter DNA strand after DNA synthesis, thus increasing the least six S. cerevisiae proteins, MSH1-MSH6, which fidelity of DNA replication. Normally, DNA in E. coli is show a high degree of amino acid similarity with the methylated at GATC sites by the Dam methylase. This Bacterial MutS proteins (Fig. 1; Table 1). Homologs of is a postreplication modification, however, and the many of these proteins have been identified in other or- daughter DNA strand is transiently unmethylated after ganisms and are the subject of intCnse study. As will be DNA replication. The MutHLS system utilizes this discussed below, there is evidence that three of these modification asymmetry to direct repair to the unmeth- proteins, MSH2, MSH3, and MSH6, function in a eu- ylated daughter strand via the activity of MutH, which karyotic MutHLS-like mismatch repair pathway. The in- nicks the unmethylated DNA strand at hemimethylated volvement of three MutS homologs in mismatch repair GATC sites. It is important to point out that the require- provides evidence for the relative complexity of eukary- ment for MutH and the use of Dam methvlation to dis- otic mismatch repair compared to the bacterial para- tinguish between newly replicated and parental DNA digm. appear to be unique to certain bacteria. Other bacteria It was orignally thought that the eukaryotic MSH2 like Streptococcus pneumoniae have a mismatch repair protein, a homolog of the bacterial MutS protein, was the system that is closely related to the MutHLS system but mismatch repair protein that recognizes mispaired bases the Streptococcus system, referred to as the Hex system, in DNA (Reenan and Kolodner 1992a,b). Purified MSH2 is not presently known to utilize a MutH homolog or from both S. cerevisiae and human cells recognizes both DNA methylation as a mechanism for strand discrimi- single-base mispairs and, with a higher affinity, multiple nation in the repair process (Modrich 1991; Fishel and base insertion/deletion mispairs (Fishel et al. 1994; Kolodner 1995; Friedberg et al. 1995; Kolodner 1995; Alani et al. 1995). More recently, it has been appreciated Modrich and Lahue 1996). DNA methylation does not that human MSH2 copurifies with a second MutS ho- occur in S. cerevisiae (Proffitt et al. 1984), and no con- molog, human GTBPIpl60 (Drummond et al. 1995; Pal- 1434 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair Table 1. Names and alternate names for mismatch repair al. 1996). Mutations in MSH2 were found to cause a genes and proteins discussed strong general mutator phenotype resulting in the accu- mulation of both frameshift and single-base substitution E. coli protein S, cerevisiae protein Human protein mutations, high rates of frameshift mutation reversion, MutS MSH2 MSH2 and dinucleotide repeat instability, consistent with a MSH3 MSH3, DUGI, MRPl general role in mismatch repair (Reenan and Kolodner MSH6 GTBP, p 160 1992a; Strand et al. 1993; Marsischky et al. 1996). In MutL PMS 1 PMS2 contrast, mutations in MSH6 caused only a modest mu- MLHl MLHl tator phenotype that was confined primarily to the ac- MLH2 PMS 1 cumulation of single-base substitution mutations, sug- General nomenclature: Gene names, uppercase italics; muta- gesting that MSH6 might be required primarily for rec- tion names, lowercase italics; protein names, uppercase except ognition of single-base substitution mispairs and to a for E. coli poteins, where only the first letter of each word is lesser extent single base insertionldeletion mispairs capitalized. (Marsischky et al. 1996). Consistent with this latter Note on MutL-related proteins: Human PMS2 is more closely view, pms3 mutants, which had been shown previously related to S. cerevisiae PMSl than human PMSl is related to S. in transformation assays that detect mismatch repair di- cerevisiae PMS1. Human PMSl is more closely related to S. rectly to cause a defect in the repair of single-base sub- cerevisiae MLH2 (see Fig. 4). stitution mispairs but not in the repair of single-base References: E. coli MutS-related proteins: (Reenan and Kolod- insertionldeletion mispairs, are caused by mutations in ner 1992a, Fishel et al. 1993; Leach et al. 1993; New et al. 1993; Drummond et al. 1995; Palombo et al. 1995; Marsischky et al. MSH6 (B. Kramer et al. 1989; Marsischky et al. 1996). 1996; Watanabe et al. 1996). E. coli MutL-related proteins: Also consistent with this view is the observation that (Bronner et al. 1994; Nicolaides et al. 1994; Papadopoulos et al. msh6 mutations cause only a small increase in dinucle- 1994; Prolla 1994a,b). otide repeat instability (Johnson et al. 1996). Unlike either msh2 or msh6 mutants, msh3 mutants have little if any mutator phenotype in forward mutation ombo et al. 1995), which appears to be the homolog of S. and frameshift reversion assays and show increased di- cerevisiae MSH6 (Figs. 2 and 3, and below). The obser- nucleotide repeat instability, albeit at lower levels, than vation that the heterodimer of human MSH2 and GTBP that seen in msh2 mutants (New et al. 1993; Alani et al. (MSH6) recognizes mispaired bases initially suggested 1994; Strand et al. 1995; Marsischky et al. 1996;). Strik- that the entire complex could function as the MutS ingly, when double mutant strains were analyzed, msh3 equivalent in eukaryotic mismatch repair (Drummond and msh6 mutations showed a large synergistic effect on et al. 1995; Palombo et al. 1995). The available data on the rate of accumulation of mutations in frameshift re- mismatch repair-defective tumor cell lines, however, version assays that measure the formation and repair of suggest that the role of a MSH2-GTBP (MSH6) complex single-base insertion1 deletion mispairs (Marsischky et in mismatch recognition may not be simple. In vivo, al. 1996). A similar synergistic effect was seen in forward tumor cell lines lacking MSH2 appear to have a high degree of microsatellite instability at both dinucleotide and mononucleotide repeat loci (Bhattacharyya et al. Mouse 1994; Shibata et al. 1994; Umar et al. 1994b; Boyer et al. Rat Human 1995; Papadopoulos et al. 1995). In contrast, tumor cell Xenopus lines lacking GTBP (MSH6) appear to have less pro- Drosophila S.cerevisiae nounced microsatellite instability: Dinucleotide repeat Human instability is difficult to detect in such cell lines, and Mouse Sxerevisiae mononucleotide repeat instability is less pronounced S.cerevisiae compared to msh2 mutant lines (Bhattacharyya et al. Human E.coli 1994; Shibata et al. 1994; Papadopoulos et al. 1995). Bio- MUTS MUTS S.typhirnurium chemical analysis also suggests that mutations in MSH2 MUTS H.influenzae and GTBP (MSH6) have somewhat different effects on MUTS A.vinlandii MUTS T.aquaticus mismatch repair in in vitro mismatch repair assays MUTS T.thermophilus (Drummond et al. 1995). Finally, mutations in MSH2 are MUTS B.subtilis HEXA S.pneurnoniae common in hereditary nonpolyposis colorectal cancer MSHI S.cerevisiae (HNPCC) families, whereas mutations in GTBP (MSH6) -I I 'J-L;:;3;::: have not been found (Papadopoulos et al. 1995; Liu et al. 1996). These data suggest that MSH2 and GTBP (MSH6) are not genetically equivalent. Genetic and biochemical analyses of mismatch repair Figure 2. Family tree of known MutS homolog proteins. All of in S. cerevisiae have suggested that mismatch recogni- the sequences used for this construction were either retrieved tion involves three MutS homologs: MSH2, MSH3, and from GenBank or are from unpublished studies; human MSH5, MSH6 (the homolog of human GTBPIpl GO), which form N. Winand and R. Kolodner (unpubl.); mouse MSH6, G. Crouse two different heterodimeric complexes (Marsischky et and R. Kolodner (unpubl.). GENES & DEVELOPMENT 1435 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press MSH2-MSH3 complexes about equally, and that repair Single-base mispair recognition of larger insertionldeletion mispairs likely utilizes the MSH2-MSH3 complex 5-10 times as frequently as the MSH2-MSH6 complex (Johnson et al. 1996; Marsischky et al. 1996). A crucial test of this model will require the purification of these two complexes and the demonstra- tion that they have the required mispair recognition properties. The genetics of S. cerevisiae MSH2, MSH3, and MSH6, Insertion-deletion mispair recognition where comparable data exist, are consistent with the re- sults of the analysis of human tumor cell lines. Like S. cerevisiae msh2 mutants, human msh2 mutant cell lines (e.g., LoVo, HEC59, 2774) have a strong general mutator phenotype and strong microsatellite instability including instability at dinucleotide and mononucle- otide repeat loci (Bhattacharyya et al. 1994; Shibata etal. 1994; Boyer et al. 1995; Liu et al. 1995). Similar to the mutator phenotype of S. cerevisiae msh6 mutants, gtbp (msh6) mutant tumor lines (DLDl/HCTlS, VAC0543) have a strong mutator phenotype when the accumula- tion of mutations at HPRT (forward mutation assay) is measured, but their microsatellite instability phenotype is reduced compared to msh2 mutant tumor lines (Bhat- Figure 3. Model for mismatch recognition in S. cerevisiae. The various postulated complexes between MSH2 and either MSH3 tacharyya et al. 1994; Shibata et al. 1994; Eshleman et al. or MSH6 are illustrated interacting with either a single-base 1995; Papadopoulos et al. 1995). Little is known about substitution mispair or an insertioddeletion mispair; exactly the possible presence of msh3 mutations in either tumor which of the proteins in these complexesMSH2, MSH3, or cell lines or HNPCC families other than the fact that MSHGactually interacts with the mispaired base is not like GTBP (MSH6) mutations they must be rare because known. Also indicated is the previously described MLHI-PMS1 most HNPCC families studied have either msh2 or mlhl complex that interacts with the mispair recognition complex. mutations (Papadopoulos et al. 1995; Liu et al. 1996). The S. cerevisiae protein names are given as primary names; the The genetic properties of MSH2, MSH3, and MSH6 in human protein names are the same except for PMS1, which is S. cerevisiae have important implications for the analy- called PMS2 in humans, and MSH6, which has been called sis of mismatch repair-defective mutations and their as- GTBP or p160 in humans. (Reprinted, with permission, from Marsischky et al. 1996). sociation with cancer susceptibility. First, the redun- dancy of MSH3 and MSH6 compared to the apparently universal requirement for MSH2 in mismatch repair pro- vides an explanation for the high prevalence of msh2 mutation assays that detect a broader variety of muta- mutations in HNPCC families compared to mutations tion events, and in these experiments, the mutation rate in either GTBP (MSH6) or MSH3 (Papadopoulos et al. and mutation spectrum in the msh3 msh6 double mu- 1995, Liu et al. 1996). This is because independent mu- tant were essentially the same as that seen in the msh2 tations in both MSH3 and GTBP (MSH6) would be re- single mutant (Marsischky et al. 1996). These and other quired to produce the same mismatch repair defect as genetic data indicate that MSH3 and MSH6 encode re- that caused by mutations in MSH2. Second, selection for dundant activities that act in MSH2-dependent mis- tumors and syndromes (HNPCC) associated with high match repair (Johnson et al. 1996; Marsischky et al. degrees of microsatellite instability may preclude can- 19961. Consistent with this hypothesis, MSH2 forms cers and syndromes caused by msh6 or msh3 mutations. heterodimeric complexes with both MSH3 and MSH6 In contrast, one might expect to find msh3 or gtbp proteins (Marsischky et al. 1996). (msh6) mutations associated with cancers and cancer The analysis of S. cerevisiae MSH2, MSH3, and MSH6 susceptibility syndromes with mutator phenotypes that has led to the proposal of a model in which there are two more closely resemble the mutator phenotypes caused different pathways of MSH2-dependent mismatch repair: by msh3 and msh6 mutations in S. cerevisiae. Third, it is repair that is primarily specific for single-base substitu- possible that the mutator phenotype caused by msh3 or tion mispairs and requires a MSH2-MSH6 complex, and gtbp (msh6) mutations compared to msh2 mutations is repair that is primarily specific for insertion/deletion insufficient to cause cancer initiation or progression or mispairs and requires either a MSH2-MSH3 complex or that such mutations have lower penetrance than msh2 a MSH2-MSH6 complex (Fig. 3; Marsischky et al. 1996). mutations. Fourth, it is possible that the lack of another The sum of the mutagenesis data reported suggests that entirely MSH2-dependent process, rather than the loss of repair of single-base substitution mispairs requires the mismatch repair, is the underlying cause of cancer sus- MSH2-MSH6 complex, that repair of single-base inser- ceptibility. Fifth, it is possible that there are tumors and tionldeletion mispairs can utilize the MSH2-MSH6 or tumor cell lines showing microsatellite instability that 1436 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair are msh3, gtbp (msh6) double mutants. A particularly karyotic mismatch repair seen in regard to the MSH pro- intriguing possibility is that mutations in MSH3 and teins. GTBP (MSH6) segregate in the human population and are In both S. cerevisiae and human cells, there appear to relatively silent until msh3 gtbp (msh6) double mutant be at least two MutL homologs that are required for mis- individuals with increased cancer susceptibility are gen- match repair: MLHl and PMSl in S. cerevisiae, and erated. It would require two independent events to inac- MLHl and PMS2 (the closest homolog of S. cerevisiae tivate the wild-type MSH3 and GTBP (MSH6) alleles as PMS1) in human cells (Prolla et al. 1994; Kolodner 1995; expected in tumor cells from HNPCC individuals (Leach Li and Modrich 1995; Modrich and Lahue 1996). These et al. 1993; Hemminki et al. 1994; Borresen et al. 1995) proteins appear to exist in a 1:l complex (Prolla et al. and the occurrence of two such events might be infre- 1994bj Li and Modrich 1995). The purified human quent. However, the loss of the second allele of one of MLH1-PMS2 complex complements the in vitro mis- the genes, MSH3 or GTBP (MSHb), would cause a weak match repair defect caused by mlhl mutations found in mutator phenotype that could speed inactivation of the human tumor cell lines, indicating that it is functional remaining functional gene. It is also possible that MSH3I in mismatch repair (Li and Modrich 1995). In S. cerevi- msh3, GTBP (MSH6)lgtbp (msh6) cells might have an siae, this MLH1-PMS1 complex interacts with MSH2 increased mutation rate resulting from unlinked partial bound to a mispaired base, consistent with the idea that noncomplementation, which would increase the loss of the MLH1-PMS1 (human PMS2) complex plays the the second alleles of these genes. same role in mismatch repair in eukaryotes that MutL plays in bacteria. The available biochemical and genetic data (discussed in more detail below) and the various Homologs of the bacterial MutL proteins protein sequence homology relationships support the idea that the S. cerevisiae MLH1-PMS1 and human A variety of studies have documented a large number of MLH1-PMS2 complexes play equivalent roles in mis- eukaryotic genes encoding proteins that show a high de- match repair. gree of amino acid similarity with the bacterial MutL In human cells there is a third MutL homolog, PMS1, proteins (Figs. 3/41. It should be noted that the complete that has been implicated in mismatch repair by virtue of gene sequence is not known in all cases. Historically, the observation of a germ-line mutation in PMSl in a these proteins have been designated as either PMS or single patient with a family history of colon cancer MLH proteins because the first eukaryotic gene encoding (Nicolaides et al. 1994). The tumor from this patient was a MutL homolog was named PMSl before it had been not examined for either loss of the second PMS1 allele or cloned. These eukaryotic proteins, however, are all re- for microsatellite instability. Therefore, it is not clear lated to the bacterial MutL proteins. The available evi- whether PMS1 behaves like other mismatch repair and dence suggests that at least two of these proteins, MLHl tumor suppressor genes during tumor development or and PMSl (PMS2 in humans), are involved in mismatch whether pmsl mutations cause a mismatch repair defect repair. The involvement of two such proteins in mis- in tumors. At present, little is known about the bio- match repair, as compared to the involvement of a single chemical role of hPMSl in mismatch repair. Additional MutL protein in bacteria, parallels the complexity of eu- studies of human PMSl are clearly needed. The genetics of S. cerevisiae PMSl and MLHl and hu- man MLHl and PMS2 parallel each other to a great ex- tent. The available data indicate that mutations in S. PMS3 Human PMSR2 Human cerevisiae MLHl and PMSl cause a similar, strong gen- PMS4 Human eral mutator phenotype that is virtually the same as that PMSB Human caused by mutations in MSH2; no differences in the mu- PMSS Human PMS2 Human tator phenotypes caused by pmsl and mlhl mutations PMSR3 Human have yet been published (Williamson et al. 1985; Bishop PMS2 Mouse PMS6 Human et al. 1987; B. Kramer et al. 1989; Strand et al. 1993; PMSR6 Human Prolla et al. 1994a; Jeyaprakash et al. 1996). Interestingly, PMSl S.cerevisiae ' PMSl Human in a recent study on the effects of msh2 msh3, and pmsl (human pms2) mutations on crossing over between di- MUTL E.coli MUTL S.typhimuria vergent DNA sequences, PMSl (human PMS2) appeared MUTL H.influenza to play significantly less of a role in suppressing homol- HEXB S.pneumonia MUTL B.subtilis ogous recombination than MSH2 (Datta et al. 1996; Hunter et al. 1996). This suggests that defects in PMS1 MLHl S.cerevisiae I lMLH1 Human I MUTL V.cholera (human PMS2) might not be equivalent to defects in other mismatch repair genes (unfortunately MLHI was Figure 4. Family tree of known MutL homolog proteins. All of not evaluated in these studies; Datta et al. 1996; Hunter the sequences used for this construction were retrieved from et al. 1996). Analysis of human tumor cell lines with GenBank except S. cerevisiae MLH2, which was provided by mutations in either MLHI or PMS2 support this view: Tom Prolla and R. Michael Liskay (Prolla 1994). Some of the sequences used (hPMS5, hPMS6, hPMSR6, hPMS8) are not Such cell lines have a strong, global microsatellite insta- complete coding sequences. bility phenotype, and extracts prepared from such cell GENES & DEVELOPMENT 1437 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner lines are mismatch repair defective (Parsons et al. 1993; progress in the identification of exonucleases that might Bhattacharyya et al. 1994; Shibata et al. 1994; Boyer et al. play a role in eukaryotic mismatch repair. 1995; Li and Modrich 1995; Risinger et al. 1995). Fur- A 5' + 3' endo/exonuclease, S. cerevisiae RTH 1 thermore, the mismatch repair defect caused by mlhl (RAD27, YKL5 10)) has been proposed to function in mis- mutations can be complemented in vitro by the addition match repair (Johnson et al. 1995). Mutations in RTHl of the MLH1-PMS2 complex (Li and Modrich 1995). In (RAD27) cause a mutator phenotype (Johnson et al. 1995; addition, cells derived from homozygous pms2 mutant Reagan et al. 1995; Vallen and Cross 1995) that in one mice show microsatellite instability, consistent with a study was shown to be as strong as that caused by mu- mismatch repair defect (Baker et al. 1995). One striking tations in MSH2, MLH1, and PMS1 (Johnson et al. 1995). feature about the genetics of the human MLHl and This result seems at odds with mechanistic studies of PMS2 genes is that there is a strikingly lower frequency mismatch repair promoted by human cell extracts indi- of germ-line pms2 mutations compared to germ-line cating that, like the E. coli reaction, eukaryotic mis- mlhl mutations in HNPCC families, especially given match repair is likely to involve redundant 5' + 3' and that mutations in both genes appear to cause the same 3' + 5' exonucleases (Modrich 1991; Fang and Modrich type of mismatch repair defect (Boyer et al. 1995; Ris- 1993; Modrich and Lahue 1996). Double mutant combi- inger et al. 1995; Liu et al. 1996). These data may reflect nations of rthl and either msh2 mlhl, or pmsl showed trivial possibilities such as founder effects (Nystrom- a three- to fivefold synergistic effect (Johnson et al. 1995). Lahti et al. 1995), or DNA sequence contexts may make One possible interpretation of this result is that RTHl MLHl more mutable than PMS2. Alternately, these data functions in an entirely different pathway from MSH2, raise the intriguing possibilities that MLHl and PMS2 do MLH1, and PMSI . Additional genetic analysis of R THl not play equivalent roles in the cell even though they are (RAD27) has indicated that RTHl (RAD27) plays a role subunits of the same complex (see above discussion of in repair of other types of DNA damage, that RTHl Datta et al. 1996; Hunter et al. 1996), that pms2 muta- (RAD27) is a member of the RAD6 epistasis group that tions confer a selective disadvantage to the population functions in DNA damage tolerance, and that rthl compared to mlhl mutations (Baker et al. 1995), or that (rad27) mutants also have a cell cycle defect (Reagan et like S. cerevisiae MSH3 and MSH6, there is a function al. 1995; Vallen and Cross 1995). Studies of the S. cere- that is redundant with PMS2. The resolution of this visiae (called RTH1, RAD27, YKL510) and human question could provide important information about (called FEN-1, MF-1, exonuclease IV) protein have sug- how mismatch repair defects give rise to cancer suscep- gested that RTHl functions in the processing of 5' ends tibility. of Okazaki fragments and processing branched DNA structures formed by different DNA repair pathways (Ishimi et al. 1988; Harrington and Lieber 1994; Waga Other components required for mismatch repair and Stillman 1994; Waga et al. 1994; Hiraoka et al. 1995; It seems clear that HNPCC families, sporadic tumors, Sommers et al. 1995). There are several possible inter- and tumor cell lines that are associated with mismatch pretations of the data on RTHl with regard to a possible repair defects or that exhibit microsatellite instability role in mismatch repair. (1) RTHl functions as an exo- exist where it has not yet been possible to demonstrate nuclease during mismatch repair; however, the strong defects in known mismatch repair genes (da Costa et al. mutator phenotype of rthl mutants seems inconsistent 1995; Katabuchi et al. 1995; Liu et al. 1995, 1996; Pap- with the mechanistic studies implicating bidirectional adopoulos et al. 1995; Wijnen et al. 1995, 1996). This excision during mismatch repair. This suggests that if observation has provided additional interest in the iden- RTHl functions in mismatch repair, it likely functions tification of other components of eukaryotic mismatch in other repair reactions that act to reduce mutation repair. At present, there is little definitive additional in- rates. (2) RTHl might not function as an exonuclease but formation about the other proteins that function in eu- rather as an endonuclease that cleaves branched struc- karyotic MutHLS-like mismatch repair pathways. On tures that must be processed during different DNA repair the basis of our current understanding of the enzymes pathways, possibly including mismatch repair. (3) The that promote MutHLS mismatch repair in E. coli, obvi- mutator phenotype seen in rthl mutants is not attribut- ous candidates for proteins required for eukaryotic mis- able to a mismatch repair defect but is rather the result match repair are exonucleases, helicases, and enzymes of defects in the fidelity of DNA synthesis, or other re- required for DNA synthesis (Modrich 199 1 Kolodner pair defects. This latter possibility could account for the 1995; Modrich and Lahue 1996). A particularly interest- modest synergistic effect of rthl mutations in combina- ing unresolved issue is the identification of the proteins tion with msh2, mlhl, and pmsl mutations. Clearly cau- and signals that function in the recognition of the newly tion must be taken in the interpretation of the meaning replicated DNA strand during mismatch repair; this is of mutator phenotypes. because DNA methylation and MutH homologs seem A second candidate for an exonuclease involved in unlikely to play this role as they do in E. coli (Modrich mismatch repair was identified in a two-hybrid screen 1991; Kolodner 1995; Modrich and Lahue 1996). with S. cerevisiae MSH2 as a bait protein (D.X. Tishkoff, Whereas most of the progress reported in eukaryotic mis- A. Beddow, M.F. Kane, and R. Kolodner, in prep.). The S. match repair enzymology has been in the area of the cerevisiae EX01 gene was identified as encoding a pro- MSH and MLH proteins, there has been some recent tein that interacts with both S. cerevisiae and human 1438 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair MSH2, and a S. cerevisiae MSH2-EX01 complex was 1995). One of several possible explanations for these data identified in vivo by coimmunoprecipitation. S. cerevi- is that there could be a separate, less-efficient repair siae EX01 appears to be a homolog of the Schizosaccha- pathway that selectively repairs insertionldeletion mi- romyces pom be 5' -+ 3' exonuclease EX0 1 (Szankasi spairs by deletion. Consistent with this view, analysis of and Smith 1992, 1995; D.X. Tishkoff, A. Beddow, M.F. mismatch repair catalyzed in extracts of human tumor Kane, and R. Kolodner, in prep.), and characterization of cell lines has suggested there may be an MLHI -indepen- the overproduced and purified S. cerevisiae EXOl has dent mismatch repair pathway that promotes the repair indicated that it preferentially degrades double-stranded of some insertionldeletion mispairs >3 nucleotides in DNA compared with single-stranded DNA. Mutations length and that repair of some such substrates may also in S. cerevisiae EX01 cause a weak general mutator phe- be MSH2-independent (Umar et al. 1994a). notype and low levels of microsatellite instability, con- More recently, several workers have detected S. cere- sistent with the redundant exonuclease proposal. Fur- visiae and human mispair binding activities present in S. thermore, analysis of double mutants was consistent cerevisiae msh2, msh3, and msh4 mutants and human with EX01 functioning in a MSH2-dependent pathway msh2 and gtbp (msh6) mutants (Miret et al. 1996; (D.X. Tishkoff, A. Beddow, M.F. Kane, and R. Kolodner, O1Regan et al. 1996). Other studies have detected activ- in prep.). S. pombe ex01 mutants also have a mutator ities that nick DNA at the site of a mispair (Chang and phenotype, and S. pombe EXOl has been proposed to Lu 1991; Yao and Kom 1994; Yeh et al. 1994). However, function in either a general MutHLS-like mismatch re- it has not been established whether these activities play pair system or a more specific mismatch repair system a role in a general mismatch repair pathway such as the like the base-specific systems described in bacteria MutHLS-like pathway or whether these activities act in (Szankasi and Smith 1995). At this stage of analysis ad- other pathways such as crossing over during meiotic re- ditional studies are clearly needed to define the potential combination, which is known to require MSH4 and roles of RTHl (RAD27) and EXOl in mismatch repair MSH5 (Ross-Macdonald and Roeder 1994; Holling- and to identify additional exonucleases that might be sworth et al. 1995), or in base-specific mismatch repair required. (Modrich 199 1 ; Kolodner 1995; Modrich and Lahue 1996). Whether these types of activities play roles in other mismatch repair pathways and whether alternate Is there another eukaryotic general mismatch repair mismatch repair pathways that recognize a broad spec- pathway? trum of mispaired bases including larger insertionldele- A number of lines of investigation have suggested that tion mispairs exist remain important unanswered ques- there may be another eukaryotic mismatch repair path- tions at this time. way that has an overlapping specificity with the MutHLS-like pathway. When gene conversion was ana- lyzed in S. cerevisiae pmsl, msh2, or mlhl mutants by Conclusions use of tetrad analysis, it was observed that there was a A considerable amount of progress has been made in the low, residual level of gene conversion independent of the identification of components of eukaryotic MutHLS-like marker analyzed (single-base substitution mutations or mismatch repair pathways. These pathways, however, small insertionldeletion mutations) (Williamson et al. appear more complex than their bacterial counterparts. 1985; Reenan and Kolodner 1992a; Alani et al. 1994; Furthermore, not all of the proteins that function in Prolla et al. 1994a). This could occur for two possible these eukaryotic pathways have been identified, nor has reasons: Either a proportion of gene conversion events it been resolved how many such repair pathways might are caused by gap repair and do not involve the formation exist. Particular challenges for the future include the of a heteroduplex recombination intermediate (discussed complete identification of the components of these re- in Alani et al. 1994), or there is an inefficient mismatch pair pathways, the understanding of how defects in these repair pathway functional in pmsl, msh2, or mlhl null pathways lead to cancer susceptibility, and the unravel- mutants. Similarly, when transformation assays were ing of the genetics of these pathways in model organ- used to analyze mismatch repair directly, not all appar- isms, which may provide greater insights into the role of ent mismatch repair was eliminated by mutations in inherited and acquired repair defects in cancer suscepti- mismatch repair genes (Bishop et al. 1987, 1989; B. bility in the human population. Kramer et al. 1989). Furthermore, the frequency of inser- tion events approximately equaled the frequency of de- letion events when insertionldeletion mispairs were re- paired after transformation of wild-type S. cerevisiae Acknowledgments cells, whereas there was a strong disparity in favor of I thank Richard Boland, Tom Kunkel, Gerry Marsischky, Dan deletion events among the fraction of insertionldeletion Tishkoff, Geoff Wahl, and Jean Wang for suggestions and com- mispairs that appeared to be repaired in a pmsl mutant ments on the manuscript. I particularly acknowledge the hos- (Bishop and Kolodner 1986; Bishop et al. 1987). A similar pitality of the Salk Institute for Biological Studies during the shift toward deletion events in mismatch repair mutants preparation of this paper. Work in my laboratory on mismatch was seen when alterations in GT tract length resulting repair has been supported by National Institutes of Health from replication errors were analyzed (Strand et al. 1993, grants GM50006 and CA6715 1. GENES & DEVELOPMENT 1439 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner References Eshleman, J.R., E.Z. Lang, G.K. Bowerfind, R. Parsons, B. Vo- gelstein, J.K.V. Wilson, M.L. Veigl, W.D. Sedwick, and S.D. Alani, E., R.A.G. Reenan, and R. Kolodner. 1994. Mismatch Markowitz. 1995. Increased mutation rate at the hprt locus repair proteins directly affect gene conversion in Saccharo- accompanies microsatellite instability in colon cancer. On- myces cerevisiae by regulating heteroduplex tract length. cogene 10: 33-37. Genetics 137: 19-39. Fang, W.-H. and P. Modrich. 1993. Human strand-specific mis- Alani, E., N.-W. Chi, and R.D. Kolodner. 1995. The Saccharo- match repair occurs by a bidirectional mechanism similar to myces cerevisiae MSH2 protein specifically binds to duplex that of the bacterial reaction. \. Biol. Chem. 268: 1 1838-1 1844. oligonucleotides containing mismatched DNA base pairs Feng, W.Y., E.H. Lee, and J.B. Hays. 199 1. Recombinogenic pro- and loop insertions. Genes & Dev. 9: 234-247. cessing of UV-light photoproducts in nonreplicating phage Anthoney, D.A., A.J. McIlwrath, W.M. Gallagher, A.R.M. Edlin, DNA by the Escherichia coli methyl-directed mismatch re- and R. Brown. 1996. Microsatellite instability, apoptosis, pair system. Genetics 129: 1007-1020. and loss of p53 function in drug-resistant tumor cells. Can- Fishel, R. and R.D. Kolodner. 1995. Identification of mismatch cer Res. 56: 1374-1381. repair genes and their role in the development of cancer. Baker, S.M., C.E. Bronner, I. Zhang, A. Plug, M. Robatzek, G. Curr. Opin. Genet. Dev. 5: 382-395. Warren, E.A. Elliott, J. Yu, T. Ashley, N. Arnheim, R.A. Fla- Fishel, R.A., M.K. Lescoe, M.R.S. Rao, N. Copland, N. Jenkins, vell, and R.M. Liskay. 1995. Male mice defective in the DNA J. Garber, M. Kane, and R. Kolodner. 1993. The human mu- mismatch repair gene PMS2 exhibit abnormal chromosome tator gene homolog MSH2 and its association with heredi- synapsis in meiosis. Cell 82: 309-320. tary nonpolyposis colon cancer. Cell 75: 1027-1038. Bhattacharyya, N.P., A. Skandalis, A. Ganesh, J. Groden, and M. Fishel, R.A., A. Ewel, S. Lee, M.K. Lescoe, and J. Griffith. 1994. Meuth. 1994. Mutator phenotype in human colorectal car- Binding of mismatched microsatellite DNA sequences by cinoma cell lines. Proc. Natl. Acad. Sci. 91: 63 19-6323. the human MSH2 protein. Science 266: 1403-1405. Bishop, D.K. and R.D. Kolodner. 1986. Repair of heteroduplex Fram, R. J., P.S. Cusick, J.M. Wilson, and M.G. Marinus. 1985. plasmid DNA after transformation into Saccharomyces cer- Mismatch repair of cis-Diamminedichloroplatinum(II)-In- evisiae. Mol. Cell. Biol. 6: 3401-3409. duced DNA damage. Mol. Pharmacol. 28: 51-55. Bishop, D.K., M.S. Williamson, S. Fogel, and R.D. Kolodner. Friedberg, E.C., G.C. Walker, and W. Siede. 1995. DNA repair 1987. The role of heteroduplex correction in gene conversion and mutagenesis. ASM Press, Washington, D.C. in Saccharomyces cerevisiae. Nature 328: 362-364. Grilley, M., J. Griffith, and P. Modrich. 1993. Bidirectional ex- Bishop, D.K., J. Andersen, and R.D. Kolodner. 1989. Specificity cision in methyl-directed mismatch repair. [. Biol. Chem. of mismatch repair following transformation of Saccharo- 268: 11830-1 1837. myces cerevisiae with heteroduplex plasmid DNA. Proc. Harrington, J. J. and M.R. Lieber. 1994. Functional domains Natl. Acad. Sci. 86: 3713-3717. within FEN-1 and RAD2 define a family of structure-specific Borresen, A.-L., R.A. Lothe, G.I. Meling, S. Lystad, P. Morrison, endonucleases: Implications for nucleotide excision repair. J. Lipford, M.F. Kane, T.O. Rognum, and R.D. Kolodner. Genes & Dev. 8: 1344-1355. 1995. Somatic mutations in the hMSH2 gene in microsatel- Hawn, M.T., A. Umar, J.M. Caretthers, G. Marra, T.A. Kunkel, lite unstable colorectal carcinomas. Hum. Mol. Genet. R.C. Boland, and M. Koi. 1995. Evidence for a connection 11: 2065-2072. between the mismatch repair system and the G2 cell cycle Boyer, J.C., A. Umar, J. Risinger, M.F. Kane, J. Lipford, J.C. Bar- checkpoint. Cancer Res. 55: 3721-3725. rett, R.D. Kolodner, and T.A. Kunkel. 1995. Microsatellite Hemminki, A., P. Peltomaki, J.-P. Mecklin, H. Jarvinen, R. Sa- instability, mismatch repair deficiency and genetic defects lovaara, M. Nystrom-Lahti, A. de la Chapelle, and L.A. Aal- in human cancer cell lines. Cancer Res. 55: 6062-6070. tonen. 1994. Loss of the wild-type MLHl gene is a feature of Bronner, C.E., S.M. Baker, P.T. Morrison, G. Warren, L.G. hereditary nonpolyposis colorectal cancer. Nature Genet. Smith, M.K. Lescoe, M. Kane, C. Earabino, J. Lipford, A. 8: 405-41 1. Lindblom, P. Tannergard, R.J. Bollag, A.R. Godwin, D.C. Hiraoka, L.R., J.J. Harrington, D.S. Gerhard, M.R. Lieber, and Ward, M. Nordenskjold, R. Fishel, R. Kolodner, and R.M. C.-L. Hsieh. 1995. Sequence of human FEN-1, a structure- Liskay. 1994. Mutation in the DNA mismatch repair gene specific endonuclease, and chromosomal location of the homologue hMLHl is associated with hereditary nonpoly- gene (FENI) in mouse and human. Genomics 25: 220-225. posis colon cancer. Nature 368: 258-261. Hollingsworth, N.M., L. Ponte, and C. Halsey. 1995. MSHS, a Chang, D.Y. and A.-L. Lu. 1991. Base mismatch-specific endo- novel MutS homolog, facilitates meiotic reciprocal recom- nuclease activity in extracts from Saccharomyces cerevi- bination between homologs in Saccharomyces cerevisiae siae. Nucleic Acids Res. 19: 4761-4766. but not mismatch repair. Genes & Dev. 9: 1728-1739. da Costa, L.T., B. Liu, W. el-Deiry, S.R. Hamilton, K.W. Kinzler, Hunter, N., S.R. Chambers, E.J. Louis, and R.H. Borts. 1996. The B. Vogelstein, S. Markowitz, J.K. Willson, A. de la Chapelle, mismatch repair system contributes to meiotic sterility in K.M. Downey, and A.G. So. 1995. Polymerase delta variants an interspecific yeast hybrid. EMBO \. 15: 1726-1 733. in RER colorectal tumours [letter]. Nature Genet. 9: 1&11. Ishimi, Y., A. Claude, P. Bullock, and J. Hurwitz. 1988. Com- Datta, A., A. Adjiri, L. New, G.F. Crouse, and S. Jinks-Robert- plete enzymatic synthesis of DNA containing the SV40 or- son. 1996. Crossovers between diverged sequences are regu- igin of replication. \. Biol. Chem. 263: 19723-19733. lated by mismatch repair proteins in yeast. Mol. Cell. Biol. Jeyaprakash, A., R. DasGupta, and R. Kolodner. 1996. Saccha- 16: 1085-1093. romyces cerevisiae pms2 mutations are alleles of MLHl and Drummond, J.T., G.-M. Li, M.J. Longley, and P. Modrich. 1995. pms2-2 corresponds to an HNPCC causing missense muta- Mismatch recognition by an hMSH2-GTBP heterodimer tion. Mol. Cell Biol. (in press). and differential repair defects in tumor cells. Science Johnson, R.E., K.K. Gopala, L. Prakash, and S. Prakash. 1995. 268: 1909-1912. Requirement of the yeast RTHl 5' to 3' exonuclease for the Eshleman, J.R. and S.D. Markowitz. 1995. Microsatellite insta- stability of simple repetitive DNA. Science 269: 238-240. bility in inherited and sporadic neoplasms. Cur. Opin. On- Johnson, R.E., G.K. Kovvali, L. Prakash, and S. Prakash. 1996. CO~. 7: 83-89. Requirement of the yeast MSH3 and MSH6 genes for MSH2- 1440 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Eukaryotic mismatch repair dependent genomic stability. J. Biol. Chem. 271: 7285-7288. Rev. Biochem. 65: 101-133. Karran, P. and M.G. Marinus. 1982. Mismatch correction of New, L., K. Liu, and G.F. Crouse. 1993. The yeast gene MSH3 06-methylguanine residues in E. coli DNA. Nature defines a new class of eukaryotic MutS homologues. Mol. & 296: 868-869. Gen. Genet. 239: 97-108. Katabuchi, H., B. van Rees, A.R. Lambers, B.M. Ronnett, M.S. Nicolaides, N.C., N. Papadopoulos, B. Liu, Y. Wei, K.C. Carter, Blazes, F.S. Leach, K.R. Cho, and L. Hedrick. 1995. Muta- S.M. Ruben, C.A. Rosen, W.A. Haseltine, R.D. Fleischmann, tions in DNA mismatch repair genes are not responsible for C.M. Fraser, M.D. Adams, J.C. Venter, M.G. Dunlop, S.R. microsatellite instability in most sporadic endometrial car- Hamilton, G.M. Petersen, A. de la Chapelle, B. Vogelstein, cinomas. Cancer Res. 55: 5556-5560. and K. Kinzler. 1994. Mutations of two PMS homologues in Kolodner, R.D. 1995. Mismatch repair: Mechanisms and rela- hereditary nonpolyposis colon cancer. Nature 371: 75-80. tionshlp to cancer susceptibility. Trends Biochem. 20: 397402. Nystrom-Lahti, M., P. Kristo, N.C. Nicoladies, S.-Y. Chang, Kramer, B., W. Kramer, M.S. Williamson, and S. Fogel. 1989. L.A. Aaltonen, A.-L. Moisio, H.J. Jarvinen, J.-P. Mecklin, K. Heteroduplex DNA correction in Saccharomyces cerevisiae Kinzler, B. Vogelstein, A. de la Chapelle, and P. Peltomaki. is mismatch specific and requires functional PMS genes. 1995. Founding mutations and Alu-mediated recombination Mol. Cell. Biol. 9: 4432-4440. in hereditary colon cancer. Nature Med. 1: 6054-6058. Kramer, W., B. Kramer, M.S. Williamson, and S. Fogel. 1989. OIRegan, N.E., P. Branch, P. Macpherson, and P. Karran. 1996. Cloning and nucleotide sequence of DNA mismatch repair hMSH2-independent DNA mismatch recognition by human gene PMSl from Saccharomyces cerevisiae: Homology of proteins. 1. Biol. Chem. 271: 1789-1796. PMSl to procaryotic MutL and HexB. 1. Bacterial. Palombo, F., P. Gallinari, I. Iaccarino, T. Lettieri, M. Hughes, A. 171: 5339-5346. D'Arrigo, 0. Truong, J.J. Hsuan, and J. Jiricny. 1995. GTBP, Lahue, R.S., K.G. Au, and P. Modrich. 1989. DNA mismatch a 160 kD protein essential for mismatch binding activity in correction in a defined system. Science 245: 160-164. human cells. Science 268: 1912-1914. Leach, F.S., N.C. Nicolaides, N. Papadopoulos, B. Liu, J. Jen, R. Papadopoulos, N., N.C. Nicolaides, Y.-F. Wei, S.M. Ruben, K.C. P. Peltomaki, P. Sistonen, L.A. Aaltonen, M. Nys- Parsons, Carter, C.A. Rosen, W.A. Haseltine, R.D. Fleischmann, C.M. trom-Lahti, X.-Y. Guan, J. Zhang, P.S Meltzer, J.-W. Yu, F.-T. Fraser, M.D. Adams, J.C. Venter, S.R. Hamilton, G.M. Pe- Kao, D.J. Chen, K.M. Cerosaletti, R.E.K. Fournier, S. Todd, tersen, P. Watson, H.T. Lynch, P. Peltomaki, J.-P. Mecklin, T. Lewis, R.J. Leach, S.L. Naylor, J. Weissenbach, J.-P. Meck- A. de la Chapelle, K.W. Kinzler, and B. Vogelstein. 1994. lin, H. Jarvinen, G.M. Petersen, S.R. Hamilton, J. Green, J. Mutation of a mutL homolog in hereditary colon cancer. Jass, P. Watson, H.T. Lynch, J.M. Trent, A. de la Chapelle, Science 263: 1625-1629. K.W. Kinsler, and B. Vogelstein. 1993. Mutations of a mutS Papadopoulos, N., N.C. Nicolaides, B. Liu, R.E. Parsons, C. Len- homolog in hereditary nonpolyposis colorectal cancer. Cell gauer, F. Palombo, A. D'Arrigo, S. Markowitz, J.K.V. Wilson, 75: 1215-1225. K.W. Kinzler, J. Jiricny, and B. Vogelstein. 1995. Mutations Li, G.-M. and P. Modrich. 1995. Restoration of mismatch repair of GTBP in genetically unstable tumors. Science 268: 1915- to nuclear extracts of H6 colorectal tumor cells by a hetero- 1917. dimer of human MutL homologs. Proc. Natl. Acad. Sci. Parsons, R., G.-M. Li, M.J. Longley, W.-H. Fang, N. Papadopou- 92: 1950-1954. los, J. Jen, A. de la Chapelle, K.W. Kinzler, B. Vogelstein, and Liu, B., M.C. Nicolaides, S. Markowitz, J.K.V. Willson, J.R. Par- P. Modrich. 1993. Hypermutability and mismatch repair de- sons, J. Jen, N. Papadopolous, P. Peltomaki, A. de la ficiency in RER + tumor cells. Cell 75: 1227-1236. Chapelle, S.R. Hamilton, K.W. Kinzler, and B. Vogelstein. Proffitt, J.H., J.R. Davie, D. Swinton, and S. Hattman. 1984. 1995. Mismatch repair gene defects in sporadic colorectal 5-Methylcytosine is not detectable in Saccharomyces cere- cancers with microsatellite instability. Nature Genet. 9: 48- Mol. Cell. Biol. 4: 985-988. visiae. Prolla, T.A. 1994. Dual requirement in yeast DNA mismatch Liu, B., R. Parsons, N. Papadopoulos, N.C. Nicolaides, H.T. repair for MLHl and PMS1, two homologs of bacterial MutL. Lynch, P. Watson, J.R. Jass, M. Dunlop, A. Wyllie, P. Pelto- Ph.D. thesis. Yale University, New Haven, CT. maki, A. de la Chapelle, S.R. Hamilton, R. Vogelstein, and Prolla, T.A., D.-M. Christie, and R.M. Liskay. 1994a. A require- K.W. Kinzler. 1996. Analysis of mismatch repair genes in ment in yeast DNA mismatch repair for MLHl and PMSI, hereditary non-polyposis colorectal cancer patients. Nature two homologs of the bacterial mutL gene. Mol. Cell. Biol. Med. 2: 169-174. 14: 407-415. Marra, G. and C.R. Boland. 1995. Hereditary nonpolyposis co- Prolla, T.A., Q. Pang, E. Alani, R.D. Kolodner, and R.M. Liskay. lorectal cancer: The syndrome, the genes and historical per- 199413. Interactions between the MSH2, MLHl and PMSl spectives. f. Natl. Cancer Inst. 87: 11 14-1 125. proteins during the initiation of DNA mismatch repair. Sci- Marsischky, G.T., N. Filosi, M.F. Kane, and R. Kolodner. 1996. ence 265: 1091-1093. Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 Rayssiguier, C., D.S. Thaler, and M. Radman. 1989. The barrier in MSH2-dependent mismatch repair. Genes & Dev. to recombination between Escherichia coli and Salmonella 10: 407-420. typhimurium is disrupted in mismatch-repair mutants. Na- Mellon, I. and G.N. Champe. 1996. Products of DNA mismatch ture 342: 396-40 1. repair genes mutS and mutL are required for transcription- Reagan, M.S., C. Pittenger, W. Siede, and E.C. Friedberg. 1995. coupled nucleotide excision repair of the lactose operon in Characterization of a mutant strain of Saccharomyces cere- Escherichia coli. Proc. Natl. Acad. Sci. 93: 1292-1297. visiae with a deletion of the RAD27 gene, a structural ho- Miret, J.J., B.O. Parker, and R.S. Lahue. 1996. Recognition of mologog of the RAD2 nucleotide excision repair gene. J. Bac- DNA insertionldeletion mismatches by an activity in Sac- teriol. 177: 364-371. charomyces cerevisiae. Nucleic Acids Res. 24: 721-729. Reenan, R.A.G. and R.D. Kolodner. 1992a. Characterization of Modrich, P. 1991. Mechanisms and biological effects of mis- insertion mutations in the Saccharomyces cerevisiae MSHl match repair. Annu. Rev. Genet. 25: 229-253. and MSH2 genes: Evidence for separate mitochondria1 and Modrich, P. and R. Lahue. 1996. Mismatch repair in replication nuclear functions. Genetics 132: 975-985. fidelity, genetic recombination and cancer biology. Annu. . 1992b. Isolation and characterization of two Saccharo- GENES & DEVELOPMENT 1441 Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press Kolodner myces cerevisiae genes encoding homologs of the bacterial L. Bisgaard, J. Mohr, R. Kolodner, and R. Fodde. 1996. Ma- HexA and MutS mismatch repair proteins. Genetics jority of hMLH1 mutations responsible for hereditary non- 132: 963-973. polyposis colorectal cancer (HNPCC) cluster at the exonic region 15-16. Am. 1. Hum. Genet. 58: 300-307. Risinger, J.I., A. Umar, J.C. Barrett, and T.A. Kunkel. 1995. A Williamson, M.S., J.C. Game, and S. Fogel. 1985. Meiotic gene hPMS2 mutant cell line is defective in strand-specific mis- conversion mutants in Saccharomyces cerevisiae. I. Isola- match repair. J. Biol. Chem. 270: 18183-18186. tion and characterization of pmsl -1 and pmsl -2. Genetics Ross-Macdonald, P. and G.S. Roeder. 1994. Mutation of a mei- osis-specific MutS homolog decreases crossing over but not 110: 609-646. mismatch correction. Cell 79: 1069-1080. Yao, M. and Y.W. Kom. 1994. Strand-specific cleavage of mis- match-containing DNA by deoxyinosine 3'-endonuclease Selva, E.M., L. New, G.F. Crouse, and R.S. Lahue. 1995. Mis- from Escherichia coli. 1. Biol. Chem. 269: 31390-31396. match correction acts as a barrier to homeologous recombi- nation in Saccharomyces cerevisiae. Genetics 139: 1175- Yeh, Y.C., H.F. Liu, C.A. Ellis, and A.-L. Lu. 1994. Mammalian 1188. topoisomerase I has base mismatch nicking activity. 1. Biol. Shibata, D., M.A. Peinado, S. Ionov, S. Malkhosyan, and M. Chem. 269: 15498-1 5504. Perucho. 1994. Genomic instability in repeated sequences in an early somatic event in colorectal tumorigenesis that per- sists after transformation. Nature Genet. 6: 273-28 1. Sommers, C.H., E.J. Miller, B. Dujon, S. Prakash, and L. Prakash. 1995. Conditional lethality of null mutations in RTHl that encodes the yeast counterpart of a mammalian 5'-to 3'-exo- nuclease required for lagging strand DNA synthesis in re- constituted systems. I. Biol. Chem. 270: 4193-4196. Strand, M., T.A. Prolla, R.M. Liskay, and T.D. Petes. 1993. De- stabilization of tracts of simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature 365: 274-276. 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. 92: 10418-10421. Szankasi, P. and G.R. Smith. 1992. A DNA exonuclease induced during meiosis of Schizosaccharomyces pombe. 1. Biol. Chem. 267: 3014-3023. . 1995. A role for exonuclease I from S. pombe in muta- tion avoidance and mismatch correction. Science 267: 1166- Umar, A., J.C. Boyer, and T.A. Kunkel. 1994a. DNA loop repair by human cell extracts. Science 266: 814-816. Umar, A., J.C. Boyer, D.C. Thomas, D.C. Nguyen, J.I. Risinger, J. Boyd, Y. Ionov, M. Perucho, and T.A. Kunkel. 1994b. De- fective mismatch repair in extracts of colorectal and endom- etrial cancer cell lines exhibiting microsatellite instability. 1. Biol. Chem. 269: 14367-14370. Vallen, E.A. and R.R. Cross. 1995. Mutations in RAD27 define a potential link between G1 cyclins and DNA replication. Mol. Cell Biol. 15: 4291-4302. Waga, S. and B. Stillman. 1994. Anatomy of a DNA replication fork revealed by reconstitution of SV40 DNA replication in vitro. Nature 369: 207-212. Waga, S., G. Bauer, and B. Stillman. 1994. Reconstitution of complete SV40 replication with purified replication factors. 1. Biol. Chem. 269: 10923-10934. Watanabe, A., M. Ikejima, N. Suzuki, and T. Shimada. 1996. Genomic organization and expression of the human MSH3 gene. Genomics 31: 311-318. Wijnen, J., H. Vassen, P. Meera Kahn, F.H. Menko, H. van der Klift, C. van Leeuwen, M. van den Broek, I. van Leeuwen- Comelisse, F. Nagengast, A. Meijers-Heijboer, D. Lindhout, G. Griffioen, A. Cats, J. Kleibeuker, L. Varesco, L. Bertario, M.L. Bisgaard, J. Mohr, and R. Fodde. 1995. Seven new mu- tations in hMSH2, an HNPCC gene, identified by denaturing gradient gel electrophoresis. Am. J. Hum. Genet. 56: 1060- Wijnen, J., P. Meera Kahn, H. van der Klift, H. Vasen, F.H. Menko, F. Nagengast, A. Meijers-Heijboer, D. Lindhout, G. Griffioen, A. Cats, J. Kleibeuker, L. Varesco, L. Bertario, M.- 1442 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on November 4, 2021 - Published by Cold Spring Harbor Laboratory Press R Kolodner Genes Dev. 1996, 10: Access the most recent version at doi:10.1101/gad.10.12.1433 This article cites 84 articles, 49 of which can be accessed free at: References http://genesdev.cshlp.org/content/10/12/1433.full.html#ref-list-1 License Receive free email alerts when new articles cite this article - sign up in the box at the top Email Alerting right corner of the article or click here. Service Copyright © Cold Spring Harbor Laboratory Press

Journal

Genes & Development – Unpaywall

Published: Jun 15, 1996

There are no references for this article.