Mismatch Repair in Human Nuclear Extracts
2002; Elsevier BV; Volume: 277; Issue: 29 Linguagem: Inglês
10.1074/jbc.m200357200
ISSN1083-351X
Autores Tópico(s)Cancer Genomics and Diagnostics
ResumoMammalian mismatch repair (MMR) systems respond to broad ranges of DNA mismatches and lesions. Kinetic analyses of MMR processing in vitro have focused on base mismatches in a few sequence contexts, because of a lack of general and quantitative MMR assays and because of the difficulty of constructing a multiplicity of MMR substrates, particularly those with DNA lesions. We describe here simple and efficient construction of 11 different MMR substrates, by ligating synthetic oligomers into gapped plasmids generated using sequence-specific N.BstNBI nicking endonuclease, then using sequence-specific nicking endonuclease N.AlwI to introduce single nicks for initiation of 3′ to 5′ or 5′ to 3′ excision. To quantitatively assay MMR excision gaps in base-mispaired substrates, generated in human nuclear extracts lacking exogenous dNTPs, we used position- and strand-specific oligomer probes. Mispair-provoked excision along the shorter path from the pre-existing nick toward the mismatch, either 3′ to 5′ or 5′ to 3′, predominated over longer path excision by roughly 10:1 to 20:1. MMR excision was complete within 7 min, was highly specific (90:1) for the nicked strand, and was strongly mispair-dependent (at least 40:1). Nonspecific (mismatch-independent) 5′ to 3′ excision was considerably greater than nonspecific 3′ to 5′ excision, especially at pre-existing gaps, but was not processive. These techniques can be used to construct and analyze MMR substrates with DNA mismatches or lesions in any sequence context. Mammalian mismatch repair (MMR) systems respond to broad ranges of DNA mismatches and lesions. Kinetic analyses of MMR processing in vitro have focused on base mismatches in a few sequence contexts, because of a lack of general and quantitative MMR assays and because of the difficulty of constructing a multiplicity of MMR substrates, particularly those with DNA lesions. We describe here simple and efficient construction of 11 different MMR substrates, by ligating synthetic oligomers into gapped plasmids generated using sequence-specific N.BstNBI nicking endonuclease, then using sequence-specific nicking endonuclease N.AlwI to introduce single nicks for initiation of 3′ to 5′ or 5′ to 3′ excision. To quantitatively assay MMR excision gaps in base-mispaired substrates, generated in human nuclear extracts lacking exogenous dNTPs, we used position- and strand-specific oligomer probes. Mispair-provoked excision along the shorter path from the pre-existing nick toward the mismatch, either 3′ to 5′ or 5′ to 3′, predominated over longer path excision by roughly 10:1 to 20:1. MMR excision was complete within 7 min, was highly specific (90:1) for the nicked strand, and was strongly mispair-dependent (at least 40:1). Nonspecific (mismatch-independent) 5′ to 3′ excision was considerably greater than nonspecific 3′ to 5′ excision, especially at pre-existing gaps, but was not processive. These techniques can be used to construct and analyze MMR substrates with DNA mismatches or lesions in any sequence context. mismatch repair MSH2·MSH6 heterodimer nucleotide(s) dithiothreitol Mismatch repair (MMR)1protein systems were originally defined by studies that elucidated roles of MutS-homolog (MSH) and MutL-homolog (MLH/PMS) proteins in DNA replication and recombination fidelity (1Buermeyer A.B. Deschenes S.M. Baker S.M. Liskay R.M. Annu. Rev. Genet. 1999; 33: 533-564Crossref PubMed Scopus (391) Google Scholar, 2Kolodner R.D. Marsischky G.T. Curr. Opin. Genet. Dev. 1999; 9: 89-96Crossref PubMed Scopus (730) Google Scholar, 3Jiricny J. Nystrom-Lahti M. Curr. Opin. Genet. Dev. 2000; 10: 157-161Crossref PubMed Scopus (230) Google Scholar). Both prokaryotic MutS and eukaryotic MSH2·MSH6 (MutSα) proteins were shown to recognize base mispairs in short linear synthetic DNA oligoduplexes, but binding to a variety of lesions, including UV light photoproducts (4Mu D. Tursun M. Duckett D.R. Drummond J.T. Modrich P. Sancar A. Mol. Cell. Biol. 1997; 17: 760-769Crossref PubMed Scopus (180) Google Scholar, 5Wang H. Lawrence C.W., Li, G.-M. Hays J.B. J. Biol. Chem. 1999; 274: 16894-16900Abstract Full Text Full Text PDF PubMed Scopus (95) Google Scholar), O6-methylguanine residues (6Duckett D.R. Drummond J.T. Murchie A.I.H. Reardon J. Sancar A. Lilley D.M.J. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 6443-6447Crossref PubMed Scopus (377) Google Scholar, 7Kat A. Thilly W.G. Fang W.-H. Longley M.J., Li, G.-M. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6424-6428Crossref PubMed Scopus (427) Google Scholar), 8-oxo-7,8-dihydroguanine (8Ni T.T. Marsischky G.T. Kolodner R.D. Mol. Cell. 1999; 4: 439-444Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar), and some base adducts (9Mello J.A. Acharya S. Fishel R. Essigmann J.M. Chem. Biol. 1996; 3: 579-589Abstract Full Text PDF PubMed Scopus (186) Google Scholar, 10Drummond J.T. Anthony A. Brown R. Modrich P. J. Biol. Chem. 1996; 271: 19645-19648Abstract Full Text Full Text PDF PubMed Scopus (283) Google Scholar, 11Li G.-M. Wang H. Romano L.J. J. Biol. Chem. 1996; 271: 24084-24088Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar, 12Swann P. Waters T.R. Moulton D.C., Xu, Y.-Z. Zheng Q. Edwards M. Mace R. Science. 1996; 273: 1109-1111Crossref PubMed Scopus (357) Google Scholar), by mammalian and yeast (8Ni T.T. Marsischky G.T. Kolodner R.D. Mol. Cell. 1999; 4: 439-444Abstract Full Text Full Text PDF PubMed Scopus (199) Google Scholar) MutSα proteins has also been reported. Mammalian cells lacking either MSH2·MSH6 or MLH1·PMS2 heterodimer proteins are resistant to agents that generate some of these lesions (7Kat A. Thilly W.G. Fang W.-H. Longley M.J., Li, G.-M. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6424-6428Crossref PubMed Scopus (427) Google Scholar, 13Branch P. Aquilina G. Bignami M. Karran P. Nature. 1993; 362: 652-654Crossref PubMed Scopus (351) Google Scholar, 14Wu J., Gu, L. Wang H. Geacintov N. Li G.-M. Mol. Cell. Biol. 1999; 19: 8292-8301Crossref PubMed Scopus (115) Google Scholar, 15Fritzell J.A. Narayanan L. Baker S.M. Bronner C.E. Andrew S.E. Prolla T.A. Bradley A. Jirik F.R. Liskay R.M. Glazer P.M. Cancer Res. 1997; 57: 5143-5147PubMed Google Scholar, 16Fink D. Zheng H. Nebel S. Norris P.S. Aebi S. Lin T.P. Nehme A. Christen R.D. Haas M. MacLeod C.L. Howell S.B. Cancer Res. 1997; 57: 1841-1845PubMed Google Scholar). The elevated UV mutagenesis seen in MMR-deficient bacterial (17Liu H. Hewitt S. Hays J.B. Genetics. 2000; 154: 503-512PubMed Google Scholar) and rodent cells (18Nara K.-i. Nagashima F. Yasui A. Cancer Res. 2001; 61: 50-52PubMed Google Scholar) suggests that MMR processing might correct photoproduct/base mispairs. Beyond such binding experiments, exogenous circular DNA substrates containing specific base mismatches and defined nicks for initiation of excision have been used to analyze MMR error corrections in mammalian cell-free extract (19Holmes J. Clark S. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5837-5841Crossref PubMed Scopus (336) Google Scholar, 20Thomas D.C. Roberts J.D. Kunkel T.A. J. Biol. Chem. 1991; 266: 3744-3751Abstract Full Text PDF PubMed Google Scholar). These studies have yielded important mechanistic findings, including roles for MutL-homolog proteins, and for other MMR accessory proteins such as proliferating cell nuclear antigen, and demonstration of MMR excision specificity. Error correction end-products have typically been assayed as restored restriction endonuclease recognition sites or reverted mutations in reporter genes (19Holmes J. Clark S. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5837-5841Crossref PubMed Scopus (336) Google Scholar, 20Thomas D.C. Roberts J.D. Kunkel T.A. J. Biol. Chem. 1991; 266: 3744-3751Abstract Full Text PDF PubMed Google Scholar). Such assays have not generally been possible with base mismatches in other sequence contexts, or with DNA lesion substrates, except in one special case where the lesion itself was removed by MMR (21Duckett D.R. Bronstein S.M. Taya Y. Modrich P. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 12384-12388Crossref PubMed Scopus (157) Google Scholar), which is not necessarily the biologically relevant response. It is difficult to prepare, in high yield and purity, a multiplicity of MMR substrates, particularly those with a single specific lesion in one strand and a defined single nick for initiation of excision in the other. Here we describe a simple procedure for preparation of MMR substrates and a general but precise MMR excision assay. We quantitatively analyzed the time courses of generation of excision gaps at specific positions in these MMR substrates in human nuclear extracts lacking exogenous dNTPs. Because of the low levels of adventitious nicks in these substrates, MMR excision was highly specific for mispaired versus homoduplex DNA, nickedversus continuous strands, and shorter versuslonger nick-mispair paths. We generated a variety of MMR substrates in high yield, by a new method: direct production of defined gaps in high copy number plasmids using a sequence-specific nicking enzyme, ligation of mismatch-creating synthetic oligomers into these gaps, and introduction of a defined nick using a second such enzyme. We used base-mismatched substrates and analyzed the end-products by restriction endonuclease assay as well, so that excision and error correction endpoints could be compared. However, the MMR excision assay procedures are applicable to DNA mismatches or lesions in any sequence context. All oligonucleotides were synthesized by MWG Biotech (High Point, NC). T4 polynucleotide kinase was purchased from Invitrogen. [γ-32P]ATP was from PerkinElmer Life Sciences, and Pfu DNA polymerase was from CLONTECH. Endonucleases AseI,AhdI, and N.BstNBI were purchased from New England BioLabs, from whom N.AlwI endonuclease was a generous gift. Plasmid pUC19X was previously derived from pUC19 by removal of all GAGTC sequences (endonuclease N.BstNBI site, see Table I from Ref. 22Wang H. Hays J.B. Mol. Biotechnol. 2000; 15: 97-104Crossref PubMed Scopus (27) Google Scholar). The base pair corresponding to the first C:G base pair in the EarI site at position 2488 of the original pUC19 sequence is designated base pair 1 in pUC19X and in its derivative vectors described here (see Table I and Fig. 1 below). Nucleotides (nt) in the sense strands are designated 1, 2, … , and the complementary antisense strand nucleotides are designated 1′, 2′, … (sense and antisense strands are defined by the pUC19 bla gene). Plasmid pUC19X was further modified, by PCR-mediated site-specific mutagenesis employing Pfu DNA polymerase, so as to remove all GGATC sequences (endonuclease N.AlwI sites), changing those at nt 699, 769′, 785, 866′, 883, 1347, 1646′, and 1668 to GGACC, GCATC, GAATC, GGATG, GCATC, GGACC, GAATC, and GGACC, respectively. In the product plasmid pUC19Y, we replaced DNA between the uniqueAatII and SapI sites by a staggered heteroduplex with a single mismatch (in boldface) made by annealing antisense strand oligomer 1 (5′-CATCGAGTCCGATGCGGATATTAATGTGACGGTAGCGAGTCGCTCTTCC) to sense strand oligomer 2 (5′-AGCGGAAGAGCGACTCGCTACCGTCACATTGATATCCGCATCGGACTCGATGACGT). Two new plasmids, each with only two GAGTC sites (underlined), separated by 32 nt on the antisense strand, were created by subsequent segregation of the strands during plasmid replication. The plasmid with an endonuclease AseI recognition site in the inserted sequence was designated pUC19AseI; the other was designated pUC19AseIC to indicate the T:A to C:G base change in the AseI recognition sequence AT(T →C)AAT. Next, the blunt-end homoduplex formed by annealing antisense strand oligomer 3 (5′-CTCGAGCACTCGATCCAAGCTT) to sense strand oligomer 4 (5′-AAGCTTGGATCGAGTGCTCGAG) was inserted into the unique SspI sites of pUC19AseI and pUC19AseIC to create a unique endonuclease N.AlwI site (underlined) on the sense strand, yielding plasmids pUC19CPD and pUC19CPDC, respectively. Plasmid pUC19CPDRev was constructed by inserting a blunt-end homoduplex, formed by annealing antisense strand oligomer 5 (5′-CTCGAGGATCTCGTGCACAGCTT) to sense strand oligomer 6 (5′-AAGCTGTGCACGAGATCCTCGAG) into the SspI site of pUC19AseI, again providing a unique N.AlwI site (underlined) but instead on the antisense strand. All modified sequences were confirmed by direct sequencing, and by restriction digestion where one or more diagnostic sites were available.Table IPlasmids and derived DNA substratesName1-apUC19X, pUC19Y, pUC19AseI, pUC19AseIC, pUC19CPD, pUC19CPDC, and pUC19CPDRev designate plasmids propagated in baceria; sCPDgap, sCPDCgap, sCPDRevgap, sCPD(a/t), sCPDC(g/t), sCPDRev(a/c), sCPD(a/t)n, sCPDC(g/t)n, and sCPDRev(a/c)n designate DNA molecules derived by treatments described under "Experimental Procedures."SizeNicking by endonuclease N.AlwI (coord.)1-bFirst base pair of theEarI site at 2488 in pUC19 and all derivatives described here are assigned coordinate 1. Nucleotide coordinates are unprimed for the sense strand (collinear with the bla gene) and primed for the antisense strand., statusGap creation and insertion of mispair via endonuclease N.BstNBIPrecursor1-ePlasmid and substrate construction from indicated precursors as described under "Experimental Procedures," except pUC19X (see Ref. 23).(Gap coord.)1-cCoordinates of the first and last nucleotides that were removed.Base pair (coord.)1-dIndicated base pair in the propagated plasmid, or base (mispair) constructed in the substrate. Coordinate refers to sense strand, e.g. A in A/T. bppUC19X1995(Multiple sites)(None)n/apUC19pUC19Y1995(No site)(None)n/apUC19XpUC19AseI1980(No site)(148′–179′)A/T (156)pUC19YpUC19AseIC1980(No site)(148′–179′)G/C (156)pUC19YpUC19CPD2002(23) closed(170′–201′)A/T (178)pUC19AseIpUC19CPDC2002(23) closed(170′–201′)G/C (178)pUC19AseICpUC19CPDRev2003(30′) closed(171′–202′)A/T (178)pUC19AseIsCPDgap1-fSubstrates used in this study.2002(23) closed(170′–201′)(Gapped)pUC19CPDsCPDCgap2002(23) closed(170′–201′)(Gapped)pUC19CPDCsCPDRevgap2003(30′) closed(171′–202′)(Gapped)pUC19CPDRevsCPD(a/t)2002(23) closed(170′–201′)A/T (178)sCPDgapsCPDC(g/t)1-fSubstrates used in this study.2002(23) closed(170′–201′)G/T (178)sCPDCgapsCPDRev(a/c)2003(30′) closed(171′–202′)A/C (179)sCPDRevgapsCPD(a/t)n1-fSubstrates used in this study.2002(23) nicked(170′–201′)A/T (178)sCPD(a/t)sCPDC(g/t)n1-fSubstrates used in this study.2002(23) nicked(170′–201′)G/T (178)sCPDC(g/t)sCPDRev(a/c)n1-fSubstrates used in this study.2003(30′) nicked(171′–202′)A/C (179)sCPDRev(a/c)1-a pUC19X, pUC19Y, pUC19AseI, pUC19AseIC, pUC19CPD, pUC19CPDC, and pUC19CPDRev designate plasmids propagated in baceria; sCPDgap, sCPDCgap, sCPDRevgap, sCPD(a/t), sCPDC(g/t), sCPDRev(a/c), sCPD(a/t)n, sCPDC(g/t)n, and sCPDRev(a/c)n designate DNA molecules derived by treatments described under "Experimental Procedures."1-b First base pair of theEarI site at 2488 in pUC19 and all derivatives described here are assigned coordinate 1. Nucleotide coordinates are unprimed for the sense strand (collinear with the bla gene) and primed for the antisense strand.1-c Coordinates of the first and last nucleotides that were removed.1-d Indicated base pair in the propagated plasmid, or base (mispair) constructed in the substrate. Coordinate refers to sense strand, e.g. A in A/T.1-e Plasmid and substrate construction from indicated precursors as described under "Experimental Procedures," except pUC19X (see Ref. 23Helig J.S. Elbing K.L. Brent R. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocol in Molecular Biology. John Wiley & Sons, New York1998: 1.76-1.78Google Scholar).1-f Substrates used in this study. Open table in a new tab Plasmids pUC19CPD, pUC19CPDC, and pUC19CPDRev were transferred into Escherichia coli strain SCS110 (dam −, endA −), which was obtained from CLONTECH. Each plasmid was prepared from 3 liters of overnight LB broth by alkaline lysis and purified by at least two rounds of isopycnic sedimentation in cesium chloride plus ethidium bromide (23Helig J.S. Elbing K.L. Brent R. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocol in Molecular Biology. John Wiley & Sons, New York1998: 1.76-1.78Google Scholar). Gapped plasmids were generated from the purified plasmids as previously described (24Wang H. Hays J.B. Mol. Biotechnol. 2001; 19: 133-140Crossref PubMed Scopus (46) Google Scholar): nicking at the closely spaced N.BstNBI sites, removal of the oligonucleotides, and purification of gapped product by chromatography on benzoylated and napthoylated DEAE-cellulose. Previously, the yield of plasmid with 22-nt gaps was 30% of input DNA (24Wang H. Hays J.B. Mol. Biotechnol. 2001; 19: 133-140Crossref PubMed Scopus (46) Google Scholar). The yield here of plasmids with 32-nt gaps was 50%, presumably because of improved binding to benzoylated and napthoylated DEAE-cellulose by the larger gap. To construct substrates for MMR studies, we annealed a 10-fold excess of oligomer 7, 5′-GCGGATATTAATGTGACGGTAGCGAGTCGCTC, to 50 μg of gapped pUC19CPDC or pUC19CPD plasmid DNA, then purified the ligated product by isopycnic centrifugation in CsCl plus ethidium bromide as previously described (24Wang H. Hays J.B. Mol. Biotechnol. 2001; 19: 133-140Crossref PubMed Scopus (46) Google Scholar). The boldface thymine (nt 178′) thus created a G/T mismatch (substrate sCPDC(g/t)) or an A/T pair (substrate sCPD(a/t)). Similarly, a 10-fold excess of oligomer 8, 5′-GCGGATATCAATGTGACGGTAGCGAGTCGCTC, was ligated to gapped pUC19CPDRev, so the boldface cytosine (nt 179′) created an A/C mismatch (substrate sCPDRev(a/c)). Endonuclease N.AlwI (25Xu Y. Lunnen K.D. Kong H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 12990-12995Crossref PubMed Scopus (57) Google Scholar), which nicks one strand of double-strand DNA 4 nt 3′ to its recognition sequence of (GGATCNNNN↓, unmethylated adenines only) was used to introduce site-specific nicks into the respective plasmids, yielding substrates sCPD(a/t)n, sCPDC(g/t)n, and sCPDRev(a/c)n (see Table I and Fig. 1). We constructed nine other substrates using oligomers similar to oligomer 7 but with one or two changes (the 5th through 17th base pairs are shown in Table II (lines 2–10; see footnote), which were ligated into gapped pUC19CPDC and processed as described above.Table IIExcision analyses of mismatched-DNA substratesExcision assay was performed by incubating 100 μg of HeLa nuclear extract with 100 ng of indicated substrate without exogenous dNTPs for 7 min at 37 °C. Extracted DNA was digested with AhdI endonuclease and annealed with a 30-mer probe just 3′ of guanine of theboldfaced G/T mismatch, collinear with the bottom strand. Quantitative measurement of signals was as described under "Experimental Procedures."2-a Indicated 5′ to 3′ sequences (lengths of 13–16 nt) in the 10 mismatched DNA substrates replace the 5th through 17th nucleotides (underlined) in oligomers, otherwise known as oligomer 7 (5′-GCGGATATTAATGTGACGGTAGCGAGTCGCTC-3′), ligated into gapped plasmid, pUC19CPDC, as described under "Preparation of Gapped Plasmid DNA and Construction of Nicked DNA Substrates." The last CT/A) substrate corresponds to pUC19CPD. Open table in a new tab Excision assay was performed by incubating 100 μg of HeLa nuclear extract with 100 ng of indicated substrate without exogenous dNTPs for 7 min at 37 °C. Extracted DNA was digested with AhdI endonuclease and annealed with a 30-mer probe just 3′ of guanine of theboldfaced G/T mismatch, collinear with the bottom strand. Quantitative measurement of signals was as described under "Experimental Procedures." 2-a Indicated 5′ to 3′ sequences (lengths of 13–16 nt) in the 10 mismatched DNA substrates replace the 5th through 17th nucleotides (underlined) in oligomers, otherwise known as oligomer 7 (5′-GCGGATATTAATGTGACGGTAGCGAGTCGCTC-3′), ligated into gapped plasmid, pUC19CPDC, as described under "Preparation of Gapped Plasmid DNA and Construction of Nicked DNA Substrates." The last CT/A) substrate corresponds to pUC19CPD. Nuclear extracts were prepared from HeLaS3 cells (purchased from the National Cell Culture Center, Minneapolis, MN) as previously described (22Wang H. Hays J.B. Mol. Biotechnol. 2000; 15: 97-104Crossref PubMed Scopus (27) Google Scholar). Standard MMR mixtures (15 μl) contained 75 fmol (100 ng) of sCPDC(g/t)n or sCPDRev(a/c)n substrate, 100 μg of nuclear extract, and 750 ng of bovine serum albumin, plus the following components at the indicated concentrations: 20 mm Tris-HCl, pH 7.6; 1.5 mm ATP; 1 mm glutathione; 0.1 mmfor each of four dNTPs; 5 mm MgCl2; and 110 mm KCl. Mixtures were incubated at 37 °C for 15 min unless otherwise indicated. Reactions were terminated by the addition of 30 μl of Stop solution (25 mm EDTA, 0.67% sodium dodecyl sulfate, and 90 μg/ml proteinase K). After further incubation of mixture at 37 °C for 15 min, DNA was extracted twice with an equal volume of phenol and precipitated with ethanol. DNA was resuspended in H2O and digested with 4 units ofAseI endonuclease and 1 μg of RNase A at 37 °C for 2 h in 15 μl of AseI digestion buffer (50 mm Tris-HCl, pH 7.9; 100 mm NaCl; 10 mm MgCl2; and 1 mm dithiothreitol (DTT)). The digested products were separated by electrophoresis in 1% agarose gel in 1× TAE buffer (40 mm Tris acetate, 2 mm EDTA). Correction of (G/T) or (A/C) mismatches to (A/T) restores a second site for AseI endonuclease, which thus cuts the products into 0.8- and 1.2-kb fragments. DNA bands were visualized by staining with ethidium bromide then imaged with UVP ImageStore7500 and analyzed using ImageQuaNT software. The repair yield equals the ratio of the summed intensities of the 0.8- and 1.2-kb fragments to the total of this sum plus the intensity of the 2.0-kb band corresponding to singly cut (uncorrected) DNA. To freeze excision gaps generated during MMR, exogenous dNTPs were omitted from a standard reaction mixture containing nicked mismatched DNA substrate or various control substrates. Incubation was at 37 °C for 7 min, unless otherwise indicated. DNA was extracted and precipitated as described under "MMR Reactions in HeLa Nuclear Extracts" and digested at 37 °C for 2 h with 4 units of AhdI endonuclease plus 1 μg of RNase A in 15 μl of AhdI digestion buffer (20 mm Tris acetate, pH 7.9; 50 mm potassium acetate; 10 mm magnesium acetate, and 1 mmDTT). After 0.5 pmol of a particular 32P-labeled oligomer probe (Table III) was added, the mixture was incubated at 85 °C for 5 min then slowly cooled down to room temperature. Annealed products were separated from free oligomers by electrophoresis in 1% agarose gels in 1× TAE buffer. After DNA bands from agarose gels were measured quantitatively, as described under "MMR Reactions in HeLa Nuclear Extracts," the gels were dried and the radioactivity in each sample was measured by phosphorimaging. Measurements of DNA bands from agarose gels were used to normalize radioactivity measurements for any variations in DNA recovery and loading.Table IIIRelative excision signals and estimated yield of excised intermediateProbe3-aSequences of probes:A, 5′-gagcgactcgctaccgtcacattaatatccgc; B, 5′-cttttcggggaaatgtgcgcggaaccccta; C, 5′-atgtatccgctcatgagacaataaccctga; D, 5′-gcggatatcaatgtgacggtagcgagtcgctc; D′, 5′-gcggatattaatgtgacggtagcgagtcgctc; E, 5′-gatcaaagaatcttcttgagatgctttttttctg; F, 5′-attcaacatttccgtgtcgcccttattccc; G, 5′-ctcggtcgttcggctgcggcgagcggtatc; H, caaataggggttccgcgcacatttccccga; J, 5′-tcagggttattgtctcatgagcggatacat; K, 5′-gggcgacacggaaatgttgaatactcatac.Coordinates3-bRefer to Fig. 1; unprimed coordinates are collinear with the sense strand of the blagene.sCPDC(g/t)n3-cCoordinates: nick 23, mismatch 178.sCPD(a/t)n3-cCoordinates: nick 23, mismatch 178.sCPDC(g/t)3-fNo strand interruptions.sCPDgapsCPDRev(a/c)n3-gCoordinates: nick, 30′; mismatch, 179′.Gap signal3-dRelative to the amount of radiolabeled probe C bound to product; determined by annealing and electrophoresis (see "Experimental Procedures").Product yield3-eEstimated by assuming thatprobe C detects 40% of input DNA.Gap signal3-dRelative to the amount of radiolabeled probe C bound to product; determined by annealing and electrophoresis (see "Experimental Procedures").Product yieldGap signal3-dRelative to the amount of radiolabeled probe C bound to product; determined by annealing and electrophoresis (see "Experimental Procedures").Product yieldGap signal3-dRelative to the amount of radiolabeled probe C bound to product; determined by annealing and electrophoresis (see "Experimental Procedures").Product yieldGap signal3-hRelative to the amount ofprobe I bound to product.Product yield % % % % % A170–2010.90360.0420.042 B114–1530.97390.052 C55–84(1.00)400.052 D170′ –201′0.0310.062 D′170′ –201′0.021 E953–9860.0420.021 F1966–19951.25500.4016 G109′ –138′0.083 H232′ –261′2.4096 I (D′)171′ –202′1.0635 J55′ –84′(1.00)33 K1975′ –1′0.166Table shows results of incubation of 75 fmol of the indicated DNA substrates with 100 μg of nuclear extract in 15 μl of MMR mixtures lacking exogenous dNTPs (or in scaled-up reactions, if several probes were used for a single substrate) for 7 min at 37 °C, extraction and precipitation of DNA, and gap analysis with indicated probes, as described under "Experimental Procedures." Signals for probes fromA to G are shown relative to signal fromprobe C with substrate sCPDC(g/t)n, and signals forprobes I, J, and K relative to the signal for probe I with substrate sCPDRev(a/c)n.3-a Sequences of probes:A, 5′-gagcgactcgctaccgtcacattaatatccgc; B, 5′-cttttcggggaaatgtgcgcggaaccccta; C, 5′-atgtatccgctcatgagacaataaccctga; D, 5′-gcggatatcaatgtgacggtagcgagtcgctc; D′, 5′-gcggatattaatgtgacggtagcgagtcgctc; E, 5′-gatcaaagaatcttcttgagatgctttttttctg; F, 5′-attcaacatttccgtgtcgcccttattccc; G, 5′-ctcggtcgttcggctgcggcgagcggtatc; H, caaataggggttccgcgcacatttccccga; J, 5′-tcagggttattgtctcatgagcggatacat; K, 5′-gggcgacacggaaatgttgaatactcatac.3-b Refer to Fig. 1; unprimed coordinates are collinear with the sense strand of the blagene.3-c Coordinates: nick 23, mismatch 178.3-d Relative to the amount of radiolabeled probe C bound to product; determined by annealing and electrophoresis (see "Experimental Procedures").3-e Estimated by assuming thatprobe C detects 40% of input DNA.3-f No strand interruptions.3-g Coordinates: nick, 30′; mismatch, 179′.3-h Relative to the amount ofprobe I bound to product. Open table in a new tab Table shows results of incubation of 75 fmol of the indicated DNA substrates with 100 μg of nuclear extract in 15 μl of MMR mixtures lacking exogenous dNTPs (or in scaled-up reactions, if several probes were used for a single substrate) for 7 min at 37 °C, extraction and precipitation of DNA, and gap analysis with indicated probes, as described under "Experimental Procedures." Signals for probes fromA to G are shown relative to signal fromprobe C with substrate sCPDC(g/t)n, and signals forprobes I, J, and K relative to the signal for probe I with substrate sCPDRev(a/c)n. To test their ability to be converted into corrected products, putative gapped intermediates were generated and purified as above (before treatment with RNase A and AhdI endonuclease) and incubated with 2 units of DNA pol I Klenow fragment (Invitrogen) plus all four dNTPs (17 μm each) in 15 μl of AseI digestion buffer (see above) for 20 min at room temperature. The gap-filling reaction was terminated by heating at 85 °C for 20 min; after cooling, 4 units of AseI endonuclease and 1 μg of RNase A were added to the mixture and the incubation continued at 37 °C for 2 h. Digestion products were separated by 1% agarose gel electrophoresis and analyzed as described under "MMR Reactions in HeLa Nuclear Extracts"; product yields were compared with those obtained from identical HeLa extract MMR reactions using standard reaction conditions. Plasmids pUC19CPD, pUC19CPDC, and pUC19CPDRev (Table I), which all contain two tandem sites for the sequence-specific nicking endonuclease N.BstNBI, separated by 32 nt on the antisense strand, were used to produce gaps into which mismatch-creating oligomers were ligated (Tables I and II). Nicking at respective sites for a recently described second such enzyme, endonuclease N.AlwI (25Xu Y. Lunnen K.D. Kong H. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 12990-12995Crossref PubMed Scopus (57) Google Scholar), provides points for initiation of 5′ to 3′ or 3′ to 5′ MMR excision along the shorter paths toward the mismatches (Table I and Fig.1). Beginning with 400 μg of plasmid DNA, we routinely obtain 200 μg of gapped plasmid. Typically, ligation of 50 μg of gapped plasmid with excess oligomer yields 20–25 μg of purified product. In substrate sCPDC(g/t)n, 3′ to 5′ excision of the sense strand along the shorter nick-mispair path removes the guanine nucleotide at nt 178, and DNA resynthesis restores a recognition site for endonuclease AseI. In substrate sCPDRev(a/c)n, 5′ to 3′ excision of the antisense strand removes the cytosine at nt 179, and resynthesis again restores the AseI site. We compared time courses of MMR error correction and excision in HeLa extracts, with or without exogenous dNTPs. The substrate sCPDC(g/t)n contains a G/T mispair at bp 178/178′, and a nick 3′ to the G, at nt 23 (Table I). In repeated experiments, yields of corrected product, now cleavable by AseI endonuclease at both bp 178 and 1381, reached plateau levels of 30–45% of input substrates after 8–10 min (Fig. 2, A and D,open circles). Product yields remained constant for up to 30 min (data not shown). MMR excision, assayed as the appearance of single-strand DNA bound by radiolabeled probe A and collinear with the nicked strand at nt 170–201 (Table III and Fig. 1), preceded error correction by 50 s at half-maximal y
Referência(s)