Purified Human MSH2 Protein Binds to DNA Containing Mismatchednucleotid&

[CANCERRESEARCH54,5539â€”5542,November1,19941 Advances in Brief

Purified Human MSH2 Protein Binds to DNA Containing MismatchedNucleotid&

Richard Fishel,2 Amy Ewel, and Mary Kay Lescoe

Department ofMicrobiology and Molecular Genetics, Markey Centerfor Molecular Genetics, University of Vermont School ofMedicine, Burlington, Vermont 05405

Abstract context of the mutator hypothesis, it is proposed that mutator genes exist which increase the mutation rate of a preneoplastic cell, allowing The human hMSH2 protein is a member of a highly conserved family multiple mutations to accumulate (5). A role for hMSH2 in alteration of postreplication mismatch repair components found from bacteria to of mutation rates is clear if its biochemical functions are similar to the humai@. Alterafions of the gene coding for this protein cosegregate with, bacterial and yeast mismatch repair proteins. and are the likely cause of, chromosome 2.Iinked hereditary nonpolyposis colon cancer. PostrepUcadon mismatch repair has been found to faithfully Based on its similarity to the bacterial MutS protein (5, 27, 28), it replace misincorporated nucleotides, thereby increasing the overall fidel is anticipated that hMSH2 is responsible for the initial recognition of Ity of DNA replication. Loss of postreplicadon mismatch repair function a mismatched DNA substrate. In this report, we demonstrate that leads to a mutator phenotype, which is proposed to account for the purified hMSH2 binds specffically to mismatched nucleotides, multiple mutations required for multistep carcinogenesis. Although the thereby providing a target for the mismatch repair process. fUnctions of hMSH2 can be anticipated based on Its similarity to well characterized bacterial and yeast proteins, proof of its functions has not Materials and Methods been established. Here we demonstrate that purified hMSH2 binds specifically to mismatched nucleotides, providing a target for the excision Chemicals and Enzymes. Ultrapure Tris (acid and base), EDTA, MgCI2, repair processes characteristic of postreplication mismatch repair. MgSO4, NaC1,phenylmethylsulfonylfluoride, and analyticalgrade sodium citrate, KC1,potassium phosphate monobasic (KH2PO4),and potassium phos Introduction phatedibasic(K2HPO4)wereobtainedfromAmresco(Solon, OH). Ultrapure glycerol was obtained from Mallinckrodt,Inc. (Paris, KY). Pepstatinand HNPCC3 may affect up to 1 in 200 people in industrialized nations leupeptinwerepurchasedfromSigmaChemicalCo. (St. Louis,MO).AlP was (1â€”4).Fourgenes have been identifiedthatcosegregatewith, andare purchasedfrom PharmaciaLKB Biotechnology, Inc. (Stockholm, Sweden). the likely cause of HNPCC susceptibility in at least 50% of the Restriction endonucleases, polynucleotide kinase, and Factor Xa were identified kindredS. Within these kindreds, hMSH2 on chromosome purchasedfromNew EnglandBiolabs, Inc. (Beverly, MA). DTT and isopro pylthio-@3-galactosidewaspurchasedfromGIBCO-BRL(Gaithersburg,MD). 2p21â€”22appears to account for 60% of the cases (5), hMLHi on Purification of hMSH2 PrOtein. The overproduction of hMSH2 was chromosome3p21 appearsto accountfor 30% of the cases (6), and achieved by insertion of a recombinant hMSH2 complementary DNA se PMS1 on chromosome 2q31â€”33and PMS2 on Chromosome 7p2l quenceconsistingof the entirehMSH2gene (5) with an initiatormethionine appearsto each accountfor 5% of the cases(7). (containing an engineered NcoI site), six histidine residues, and a factor Xa hMSH2, hMLH1, hPMS1, and hPMS2 are members of two highly cleavage site (IleGluGlyArg)fused to the normalhMSH2NH2-terminalme conserved families of postreplication mismatch repair genes, MutS thionine and cloned into the NcoI site of the pET1id expression vector and MaiL, that are involved in increasing the fidelity of replication by (Novogen, Madison,WI). Inductionwith 0.1 M isopropylthio-(3-galactoside specific repair of DNA polymerase misincorporation errors (8, 9). resulted in expression of hMSH2 (Mr 106,000) in pLysS bacteria. Four liters Ablation of MutS, MutL, or their homologues in bacteria and yeast of induced bacterialcells were pelleted and washed with 150 ml 20 mM increases the rate of spontaneous mutation resulting in a mutator potassium phosphate(pH 8), 150 mMNaCl, 10 mM EDTA plus protease inhibitors (0.5 mM phenytmethylsulfonyl fluoride, 0.8 @g/mlpepstatin, 0.8 phenotype (10â€”12).The most commonly recognized mutator pheno @xg/mlleupeptin), then washed twice with 150 ml 20 mM potassium phosphate type is exhibited as replication errors of simple repetitive DNA that (jH 8), 150 mMNaCl plus proteaseinhibitorsandthenresuspendedin 20 ml results in microsatellite instability (i3â€”15).A similar microsatellite of thesamebuffer.Thesuspensionwas frozenandthawedon ice to inducecell instability mutator phenotype has been observed in tumors from lysis, and the cell debri was pelleted by centrifugation at 100,000 X g (Fraction HNPCC patients (16), a wide variety of sporadic tumors (15, 17â€”20), I, 220 mg). To thesupernate,saturatedammoniumsulfatewas addedon ice to and several cell lines derived from tumors (21). A proposed mecha a final concentrationof45%(v/v), andthe solutionwas allowed to stiron ice nism for the microsatellite instability mutator phenotype involves for an additional60 min. The precipitatewas collected by centrifugationand misrecognition of replication-induced errors that result in mismatched resuspendedin4 ml of 20 mMpotassiumphosphate(jH 8), 300 mMNaC1plus nucleotides within these repeat sequences (22). These mismatched proteaseinhibitorsand dialyzed 8 h against200 volumes of the same buffer nucleotides are subsequently fixed into the genome by a second round (Fraction II, 46 mg). Fraction II was applied to a Ni/NTA agarose column (0.5 of replication,resultingin localized genome instability. cm X 5 cm; Qiagen, Chatsworth, CA) equilibrated 20 mMpotassium phos phate(pH 8), 300 mMNaClplus proteaseinhibitors.Thecolumnwas washed A mutator phenotype has been proposed to account for the multiple with 10 volumes of the equilibration buffer and developed with pH step mutations required for multistage carcinogenesis (23â€”26).Within the gradients (pH 7, pH 6, pH 5.5, pH 4.5, and pH 4). hMSH2 eluted in the pH 4.5 step (Fraction III, 2 mg). Fraction III was diluted two fold with 20 mM Received 9/13/94; accepted 9/19/94. potassium phosphate (pH 8), 2 m@tDTT, 0.2 m@iEDTA plus protease inhib Thecostsof publicationofthisarticleweredefrayedinpartbythepaymentofpage itors and applied directly to a (0.5 cm X 8 cm) hydroxyappetite column charges.Thisarticlemustthereforebeherebymarkedadvertisementinaccordancewith 18U.S.CSection1734solelyto indicatethisfact. (Bio-Rad,Hercules,CA) equilibratedin20 mMpotassiumphosphate,100 mM @ I work was supported by NIH Grant CA 56542 and the Lake Champlain Cancer NaCl, i mMDTF, 0.1 mMEDTA plus protease inhibitors(buffer A). The ResearchOrganization. column was washed in buffer A except the phosphate concentration was 2 To whom requests for reprints should be addressed, at Department of Microbiology increasedto50 mM,followedby a lineargradientto0.25 Mphosphate.hMSH2 and Molecular Genetics, University of Vermont College of Medicine, Stafford Hall, Burlington,VT05405-0068. eluted in a broad band between 0.2 and 0.25 Mphosphate (Fraction IV, 0.7 3 The abbreviations used are: HNPCC, hereditary nonpolyposis colon cancer; D'IT, mg). Fraction IV was diluted 2-fold with buffer A minus the potassium dithiothreitol; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis. phosphateandapplieddirectlyto a (0.5 cm X 6 cm) heparin-Sepharosecolumn 5539

AB C 1 23456 50 ,@

200â€” 40@ Fig. 1. Purification and mismatch binding activ ity of hMSH2. A. Silver-stained SDS-PAGE of Frac tion V hMSH2 protein (see â€œMaterialsandMeth 116â€” odsâ€•).B, mismatch binding gel-shift assay using 97â€” Fraction V hMSH2. Lane 1, no protein; Lane 2, 24 30@ ng hMSH2;Lane 3, 74 ng hMSH2;Lane4, 239 ng counts 67â€” hMSH2; Lane 5, 399 ng hMSH2; Lane 6, 798 ng I mm2 hMSH2. C. quantitationof the binding activity kDa (x103) 20 shown in (B). Countsper mm2of the shifted mate 45â€” rialwas determinedusingthe MolecularAnalyst2.0 software associated with the BiOradphosphoimager (Bio-Rad). 10 31 â€”

@ 21 â€” 0 â€¢ â€¢ I I I I 0 200 400 600 800 MSH2(ng)

(Pharmacia) equilibrated in buffer A, washed with 10 column volumes, and tains a G-T mismatch at position 19 within the context of a 39-mer eluted with 0.1 MNaCl step gradients. hMSH2 eluted in the 0.5 MNaC1step oligonucleotide, which may be bound by a mismatch binding proteinand (Fraction V, 0.3 mg) and was homogeneous as judged by SDS-PAGE visualized by its retardation relative to the unbound material following following silver stainingfor protein(Fig. IA). gel electrophoresis (29). The binding of purified hMSH2 to the G-T Mismatch Binding Assay. Complementary strands of a 39-mer were syn mismatch-containing oligonucleotide is shown in Fig. 1B, and quantita thesized (Applied Biosystems 392, Foster City, CA). The top strand always contained the sequence 5'-TGG GCG CFA GAG GCG TGG GAA AGA GTG lion of this binding activity is shown in Fig. 1C. The binding appears to CCCTGA CCATCC.The bottomstrandforthe homoduplexoligonucleotide be cooperative and reaches 80% saturation at a protein:DNA ratio of was 5'-GGA TGG TCA 000 CAC TCT UC CCA CGC CFC TAG CGC approximately 4:1. These results are qualitatively similar to those found CCA, and for the heteroduplexoligonucleotide,it was 5'-GGA TOG TCA with the bacterial MutS protein, with the exception that hMSH2 appears GGGCAC TCT 1TF CCA CGC CrC TAG CGCCCA. The top strandwas more stable over long incubation times and it is capable of binding labeled with polynucleotide kinase and [a-32P]ATP according to the manu mismatched nucleotides with the hexa-His tag attached. facturer's recommendations. Unincorporated label was removed by DEAE Mismatch binding was sensitive to proteinase K and insensitive to chromatography, and the labeled oligonucleotide precipitated with ethanol. RNase A (data not shown), suggesting that a protein component was The labeledoligonucleotidewas thenannealedwith a complementarypartner responsible for the binding activity. We additionally found that AlP at a molar ratio of 5:1 (unlabeled complement:labeled oligonucleotide) at 55Â°C stimulates binding, allows the reaction to reach saturation at lower for 2 h. Unannealed single-stranded material was removed by treatment with BND cellulose (Sigma), and the annealed material was desalted and filtered hMSH2 concentrations,and produces more of a slower migrating with a Centri-Sep Spin Column (Princeton Separations, Adelphia, NJ). bound form that appears to be a homopolymeric multimer.5 Mismatch binding was performed in 30 @lcontaining20 m@tpotassium Proteolytic removal of the hexa-His tag or purification of untagged phosphate (pH 8), 50 mM NaCl, 1 mM Dli.', 0.1 mM EDTA, and 23% glycerol hMSH2 does not affect its activities qualitatively or quantitatively andincubatedat23Â°Cfor10 mm.The bindingreactionwas loadedonto a 4% (data not shown; see below). Furthermore, placement of the G-T acrylamide gel (40:1; acrylamide:bis; Ref. 29) and electrophoresed at 35 mA mismatch at four other sites within the 39-mer does not signfficantly for 2 h. Bound material was quantitated with a Bio-Rad Phosphoimager affect hMSH2 binding activity, suggesting the lack of any DNA (Hercules, CA). sequence context effects in hMSH2 mismatch binding. We also did not detect either a qualitative or quantitative difference in binding Results C-A, G-G, or insertion/deletion mismatches. Finally, hMSH2 binds As a first step toward understanding the function of hMSH2 in very poorly to single-stranded DNA. However, as a precaution, the human cells and correlating alterations in function with the develop annealed oligonucleotide DNA substrates were treated with BND ment of tumors in HNPCC kindreds, we have purified the normal cellulose to ensure that any remaining single-stranded oligonucleotide protein from a bacterial overproduction system (Fig. IA). We recon was eliminated from the binding reaction. structed the hMSH2 gene such that it contained a hexa-HIS tag Two experimentshave been performedto confirm hMSH2 as the consisting of an initiator methionine, six histidine residues, and a protein responsible for the mismatch binding activity. The mistn@stch Factor Xa cleavage site fused to the amino terminal methionine of the binding activity can be specifically eliminated by preequiibrati6ii and human MSH2. An initial metal affinity purification step based on centrifugation of the purified protein with Ni/NTA agarose, which hexa-HIS interaction with immobilized nickel (Ni/NTA; Qiagen, removes hexa-HIS tagged proteins via interaction between the nickel Chatsworth, CA) was then followed by two subsequent chromato metal and the histidine residues (Fig. 2, Lane 2). Furthermore, the graphic steps to ensure purification of the hMSH2 protein to apparent mismatch binding activity was left intact by treating the purified homogeneity (Fig. IA) and will be described in detail elsewhere.4 protein with Factor Xa prior to NiINTA treatment (Fig. 2, Lane 1) or A mismatch binding gel-shift assay was developed to assess restored by elution of the tagged protein from the NiINTA beads hMSH2 protein activities (30). The mismatch binding substrate con (bound in Fig. 2, Lane 2) following incubation with pH 4 buffer (Fig.

4 R. Fishel, A. Ewel, and M. K. Lescoe, Purification of human MSH2 protein, 5 R. Fishel, A. Ewel, S. Lee, M. K. Lescoe, and J. D. Griffith, Multiple mismatch manuscript in preparation. binding forms of the human MSH2 protein, manuscript in preparation. 5540

Ni I NTA + + + purified human MSH2. These results also demonstrate that the pres (pH4) ence of hexa-His tag does not significantly affect mismatch binding Factor Xa + - activity. Finally, size exclusion chromatography suggests that a pro

@ p. â€¢ tein with a relative molecular weight of approximately Mr 106,000 Â±6,000 is associated with this binding activity (data not shown). These results further implicate hMSH2 (Mr 106,400) as the protein that we have purified and which binds to DNA-containing mismatched nucleotides. To test the specificity of binding, we have carried out a competition experiment in which either unlabeled homoduplex oligonucleotide DNA (containing the identical DNA sequence except that the mis match site contained fully paired nucleotides) or unlabeled heterodu plex (G-T mismatch-containing)DNA was included in the binding reaction (Fig. 3). These data indicate that 50- to 100-fold greater homoduplex DNA competitor is required to compete hMSH2 binding from the mismatch DNA substrate than unlabeled heteroduplex com petitor. We have also tested the effect of exogenous AlP on homoduplex competition and fmd that approximately 2- to 4-fold more homoduplex DNA is required to obtain similar competition of mismatch binding, suggesting that the specificity of hMSH2 binding to mismatched DNA is increased in the presence of AlT (datanot shown). Fig. 2. Purified hMSH2 contains the eleven amino acid tag introduced into the engineered expression construct. An 11-amino acid tag consisting of an initiator methi Discussion oninc, six histidine residues, and a Factor Xa cleavage site (lleGluGlyArg) was fused to theNH@-tcrminalsnethionineofhMSH2.TreatmentofthepurifiedproteinwithFactorXa We have purifiedhMSH2 to apparenthomogeneityandfound that (NewEnglandBiolabs,Beverly,MA)was foundto releasethe tag from the normal hMSH2andmakeitresistanttospecificbindingbyNi/NTAagarose,whichinteractswith it binds to mismatched nucleotides. The protein that we have purified polyhistidinc residues (Lane 1). Tagged protein was bound to NWNTAat pH 8 (Lane 2) was judged to be hMSH2 by two criteria: (a) the purified protein andelutedatpH4 (Lane3), accordingtothemanufacturer'srecommendations(Qiagen). migrates at the expected molecular weight in an SDS-PAGE system; and(b) the purifiedproteinwas shown to containthe recombinant11 2, Lane 3). Taken together, these results suggest that the mismatch amino acid tag at the NH2-terminus that was engineered into the binding activity has residues that allow it to bind NiINTA and a Factor hMSH2 expression construct. In addition, we have found that the Xa cleavage site that is capable of removing those residues. Since it mismatch binding activity elutes as a protein with a relative molecular is highly unlikely that a mismatch binding activity, other than the weight of 106,000 in size exclusion chromatography (data not shown). recombinant hMSH2 constructed for these studies, has these com Taken together, these data suggest that we have purified hMSH2 to bined characteristics, these experiments strongly imply that we have apparent homogeneity.

A homoduplex heteroduplex Competitor (â€œ9) 40020010050 40020010050 @ Protein (ng) 0 244080239399 399r' â€¢ 399 S

Fig. 3. Competitionof mismatchbindingwith honsoduplexandheteroduplexDNA.A, titrationof mismatchbindingwith purifiedhMSH2,followed by competitionwithhomoduplexandheteroduplex DNA. The binding reaction was performed as de scribedin â€œMaterialsandMethods.â€•Afterbinding, the shownquantityofcompetitorwasintroduced intotheseactionandincubatedforan additional10 him. B. quantitation of bound material in counts per @2was performed using the Molecular Analyst SoftwareassociatedwiththeBio-Radphoshoimager @o-Ra4 B

@ counts] (xIO)

Competitor (ng) 5541

We have shown thatthe purifiedhMSH2 proteinbinds to a model 4. Peltomaki,P., Aaltonen,L. A., Sistonen,P., Pylkkanen,L., Mecklin,J-P.,Jarvinen, H.,Green,J.S., Jass,J.R.,Weber,J.L, Leach,F.S., Petersen,G.M.,Hamilton, oligonucleotide DNA containing a mismatched nucleotide. The S.R.,delaChapelle,A.,andVogelstein,B.Geneticmappingofalocuspresisposing specificity of mismatch binding was tested by competition with un to humancolorectalcancer.Science (@VashingtonDC),260: 810â€”812,1993. labeled homoduplex or heteroduplex DNA. We found that 50- to 5. Fishel,R.A., Lescoe,M. K., Rao,M. R. S., Copland,N.,Jenkins,N.,Garber,J., Kane, M., and Kolodner, R. The human mutator gene homolog MSH2 and its 100-fold more homoduplex DNA was required to effectively compete associationwithhereditarynonpolyposiscoloncancer.Cell,75: 1027â€”1038,1993. binding when compared with unlabeled heteroduplex DNA competi 6. Bromer, C. E., Baker, S. M., Morrison,P. T., Warren,G., Smith, L G., Lescoe, tor. These experiments demonstrate that hMSH2 vastly prefers to bind M. K.. Kane, M., Earabino,C., Lipford,J., Lindblom,A., Tannergard,P., Bollag, R. J., Godwin, A. R., Ward, D. C., Nordenskjold, M., Fishel, R., Kolodner, R., and DNA containing a mismatched nucleotide over simple duplex DNA. Liskay, R. M. Mutation in the DNA mismatch repair gene homologue hMLH1 is Titration of hMSH2 with mismatched DNA suggests that saturation associated with hereditary nonpolyposis colon cancer. Nature (Land.), 368: 258â€”261, 1994. occurs at a protein:DNA ratio of 4:1. In addition, a â€œsuper-shiftâ€•band 7. Nicolaides, N. C., Papadopoulos, N., Liu, B., Wei, Y-F., Carter, K. C., Ruben, S. M., becomes apparent at these saturating concentrations of hMSH2. This Rosen,C. A., Haseltine,W.A., Fleischmann,R.D.,Fraser,C. M.,Adams,M.D., band appears to represent a homopolymeric multimer of hMSH2 Venter,J. C., Dunlop, M. G., Hamilton,S. R., Petersen,G. M., de la Chapelle,A., Vogelstein,B., andKinzler,K. W. Mutationsof two PMS homologuesin hereditary bound to the mismatched oligonucleotide.5 A similar super-shift has nonpolyposiscoloncancer.Nature(Land.),371:75â€”80,1994. been observed when the yeast MLH1 and PMS1 proteins are added to 8. Modrich, P. Methyl-directed DNA mismatch correction. J. Biol. Chem., 264: yeast MSH2 bound to a mismatched DNA substrate (31). These 6597â€”6600,1989. 9. MOdriCh,P. Mechanisms and biological effects of mismatch repair. Annu. Rev. findings are the basis of a suggestion by Liskay and colleagues (31) Genet.,25: 229â€”253,1991. that a ternary complex is formed containing MSH2-MLH1-PMS1. 10. Cox, E. C. Mutatorgene studies in Escherichia coli: the mutTgene. Genetics, 73(Suppl.): 67â€”80,1973. Our results would propose an alternative possibility in which the 11. Cox, E. C. Bacterialmutatorgenes and the controlof spontaneousmutation.Annu. super-shift observed with the yeast proteins is merely an oligomeric Rev.Genet.,10: 135â€”156,1976. form of MSH2 and that MLH1 and PMS1 promote its formation at 12. Rydberg, B. Bromouracil mutagenesis and mismatch repair in mutator strains of Escherichia coli. Mutat. Rca., 52: 11â€”24,1978. lower MSH2 protein concentrations. Such a hypothesis cannot be 13. Levinson, G., and Gutman, G. A. High frequencies of short frameshifts in poly ruled out since the physical presence of the MLH1 and PMS1 proteins CAITG tandem repeats borne by bacteriophage M13 in Escherichia coli K-12. in the super-shiftcomplex was not directly determinedin the yeast Nucleic Acids Res., 15: 5313â€”5338,1987. 14. Strand,M., Prolla,T. A.. Liskay,R. M., andPetes,T. D. Destabilizationof tractsof studies. simple repetitive DNA in yeast by mutations affecting DNA mismatch repair. Nature The mismatch binding activity appears similar to both the bacterial (Land.), 365: 274â€”276,1993. 15. Ionov, Y., Peinado, M. A., Malkbosyan, S., Shibata, D., and Perucho, M. Ubiquitous and the yeast proteins, with the exception that the human protein somatic mutations in simple repeated sequences reveal a new mechanism for colonic appears to be more stable and it is capable of binding mismatched carcinogenesis.Nature(Land.),260: 558â€”561,1993. nucleotides as a tagged version. Our results are consistent with a 16. Aaltonen,L A., Peltomaki,P., Leach, F., Sistonen,P., Pylkkanen,S. M., Mecklin, J-P.,Jarvinen,H.,Powell,S.,Jen,J.,Hamilton,S.R.,Petersen,G.M.,Kinzler,K.W., proposed role for hMSH2 in the initiation of mismatch repair in Vogelstein,B., andde la Chapelle,A. aues to thepathogenesisof familialcolorectal human cells by providing a target for the excision repair process. If cancer. Science (Washington DC), 260: 812â€”816,1993. one assumes that mismatched nucleotides are an intermediate in 17. Thibodeau, S. N., Bren, G., and Schaid, D. Microsatellite instability in cancer of the proximal colon. Science (Washington DC), 260: 816â€”819, 1993. microsatallite instability, then a direct role for hMSH2 in these pro 18. Han,H-J.,Yanagisawa,A., Kato,Y., Park,J-G.,andNakamura,Y.Geneticinstability cesses can also be postulated. However, the bacterial MutS-dependent inpancreaticcancerandpoorlydifferentiatedtypeofgastriccancer.CancerRes.,53: 5087â€”5089,1993. mismatch repair system has been found to poorly repair insertion! 19. Gao,X., Wu,N., Grignon,D., Zacharek,A., Liu,H., Salkowski,A., Li, G., Sakr,W., deletion mismatches of greater than three nucleotides (32). Further Sarkar, F., Porter, A. T., Chen, Y. 0., and Honn, K. V. Microsatellite instability in more, the purified MutS protein does not efficiently recognize these humanprostaticadenocarcinoma.Oncogene,in press, 1994. 20. Risinger,J.I., &rchuck,A.,Kohler,M.F., Watson,P.,Lynch,H.1., andBoyd,J. types of insertion/deletion mismatches (27, 33). A direct role for Genetic instability of microsatellites in endometrial carcinoma. Cancer Res., 53: hMSH2 in microsatellite instability has to be assessed and the binding 5100â€”5103,1993. of hMSH2 to insertion/deletion mismatched nucleotides determined. 21. Bhattacharyya, N. P., Skandalis, A.. Ganesh, A.. Groden, 3., and Meuth, M. Mutator phenotypeinhumancolorectalcarcinomacelllines.Proc.Nail.Acad.Sci.USA,91: The absence of mismatch repair function in bacteria and lower 6319â€”6323,1994. eukaryotes has been shown to increase the rate of spontaneous mu 22. Kunkel, 1. Slippery DNA and diseases. Nature, 365: 207-208, 1993. 23. Knudson,A.G. Geneticsofhumancancer.Annu.Rev.Genet, 20: 231â€”251,1986. tation up to 1000-fold (11, 12). The finding that hMSH2 binds 24. Fearon,E.R.,andVogeistein,B.A geneticmodelforcolorectaltumorigenesis.Cell, mismatched nucleotides suggests that the loss of this function may 61: 759â€”767,1990. also lead to a mutator phenotype similar to that found in bacteria and 25. Loeb,L A.Mutatorphenotypemayberequiredformultistagecarcinogenesis.Cancer Rca., 51: 3075â€”3079,1991. yeast. It is clear that an increased mutation rate could account for the 26. Renan,M. J. How manymutationsarerequiredfortumorigenesis?Implicationsfrom multiple mutations required for multistage carcinogenesis and humancancerdata.Mol. Carcinog.,7: 139â€”146,1993. implicates hMSH2 as an initiator and propagator of multistage 27. Su, S-S., and Modrich, P. Escherichia coli mutS-encoded protein binds to mismatched DNAbasepairs.Proc.NatLAced.Sci.USA,83:5057-5061,1986. carcinogenesis. 28. Reenan, R. A. G., and Kolodner, R. D. Isolation and characterization of two Saccha romyces cerevisiae genes encoding homologs of the bacterial HexA and MutS Acknowledgments mismatchrepairproteins.Genetics,132:963â€”973,1992. 29. Ausubel,F. M., Brent,R., Kingston,R. E., Moore,D. D., Scidman,J.G., Smith,J.A., We thankRichardKolodnerand LorenaKallalfor many helpfuldiscussions. andStruhl,K. DNA-proteininteractions.Curr.ProtocolsMol. Biol., 2: 12.0.1â€”12.9.4, 1992 References 30. Jiricny,J.,Su,S-S.,Wood,S.G.,andModrich,P.Mismatch-containingoligonucleo tide duplexes houndby the E. coli mutS-encodedprotein.Nucleic Acids Res., 16: 1. Bishop, T. D., and Thomas, H. The genetics of colorectalcancer.CancerSurv., 9: 7843â€”7854,1988. 585â€”604,1990. 31. Prolla,1. A., Pang,0., Alani, E., Kolodner,R. D., andLiskay,R. M. MLH1,PMS1, 2. Lynch, H. T., Smyrk, T., Watson, P., Lanspa, S. J., Boman, B. M., Lynch, P. M., and MSH2 interactionsduring the initiation of DNA mismatch repair in yeast. Lynch, J. F., and Cavalieri, i. Hereditarycolorectal cancer. Semin. Oncol., 18: Science (Washington DC), 265: 1091â€”1093,1994. 337â€”366,1991. 32. Parker, B. 0., and Marinus, M. G. Repair of DNA heteroduplexes containing small 3. Lynch, H. T., Smyrk, T. C., Watson, P., Lanspa, S. J., Lynch, J. F., Lynch, P. M., heterologous sequences in Escherichia coli. Proc. Nail. Acad. Sd. USA, 89: Cavalieri, R. J., and Boland, C. R. Genetics, natural history, tumor spectrum, and 1730â€”1734,1992. pathologyof hereditarynonpolyposiscolorectalcancer:an updatedreview. Gastro 33. Su, S-S., Lahue, R. S., Au, K. G., and Modrich, P. Mispair specificity of methyl enterology, 104: 1535â€”1549,1993. directed DNA mismatch correction in vitro. J. Biol. Chem., 263: 6829â€”6835,1988.

5542

Richard Fishel, Amy Ewel and Mary Kay Lescoe

Cancer Res 1994;54:5539-5542.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/54/21/5539

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/54/21/5539. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.