MutL traps MutS at a DNA mismatch

DNA mismatch repair (MMR) identifies and corrects errors made during replication. In all organisms except those expressing MutH, interactions between a DNA mismatch, MutS, MutL, and the replication processivity factor (β-clamp or PCNA) activate the latent MutL endonuclease to nick the error-containing daughter strand. This nick provides an entry point for downstream repair proteins. Despite the well-established significance of strand-specific nicking in MMR, the mechanism(s) by which MutS and MutL assemble on mismatch DNA to allow the subsequent activation of MutL’s endonuclease activity by β-clamp/PCNA remains elusive. In both prokaryotes and eukaryotes, MutS homologs undergo conformational changes to a mobile clamp state that can move away from the mismatch. However, the function of this MutS mobile clamp is unknown. Furthermore, whether the interaction with MutL leads to a mobile MutS–MutL complex or a mismatch-localized complex is hotly debated. We used single molecule FRET to determine that Thermus aquaticus MutL traps MutS at a DNA mismatch after recognition but before its conversion to a sliding clamp. Rather than a clamp, a conformationally dynamic protein assembly typically containing more MutL than MutS is formed at the mismatch. This complex provides a local marker where interaction with β-clamp/PCNA could distinguish parent/daughter strand identity. Our finding that MutL fundamentally changes MutS actions following mismatch detection reframes current thinking on MMR signaling processes critical for genomic stability.The DNA mismatch repair (MMR) system employs several proteins to locate and correct DNA replication errors that escape polymerase proofreading. Mutations in these proteins contribute to MMR dysfunction that is associated with carcinogenesis, such as Lynch syndrome and other diseases associated with high mutator phenotypes (1, 2). In all organisms, MMR is initiated by binding of MutS homologs to a base–base mismatch or an insertion/deletion loop (IDL), followed by ATP-dependent recruitment of MutL homologs to begin the process of repair (3, 4). Following MutL recruitment, a key event is the introduction of a nick that directs excision and resynthesis of the nascent DNA strand containing the error (5–7).In methyl-directed MMR, which occurs in Escherichia coli, the mismatch- and ATP-dependent MutS–MutL–DNA complex activates the protein MutH to nick transiently unmethylated d(GATC) sequences in the daughter strand. Notably, however, MutH is not widely conserved in prokaryotes and does not exist in eukaryotes. Recent in vitro studies of eukaryotic MMR indicate that in these MutH-free organisms, detection of a mismatch by MutS or MutSα [MutS(α)] licenses MutL(α) to interact with the processivity factor (β-clamp/PCNA), which in turn activates the latent endonuclease activity of MutL(α) to incise the daughter DNA strand on both the 3′ and 5′ sides of the error (8–11). The interaction between MutL and the β-clamp (or between MutLα and PCNA) provides the strand discrimination signal because the β-clamp (or PCNA) is loaded asymmetrically at the replication fork or at a nick in DNA (10, 12).The importance of the nicking activity of MutL homologs is highlighted by the observation that mutations that impair yeast MutLα endonuclease activity cause a significant mutator phenotype and genomic instability (11, 13, 14). Despite the well-established significance of strand-specific nicking in MMR, the mechanism(s) by which MutS and MutL assemble on mismatched DNA to allow subsequent activation of MutL endonuclease activity by β-clamp/PCNA remains elusive. There is general agreement that in both prokaryotes and eukaryotes, after binding a mismatch MutS or MutSα can undergo conformational changes to a mobile clamp state that can move away from the mismatch (6, 15). What happens after this step is mired in controversy. Several disparate models for MutS(α)–MutL(α) mismatch complex formation and the subsequent signaling of repair have been proposed (e.g., see refs. 6, 7, 15–21). One prominent model in the field has MutL(α) joining MutS(α) to form MutS(α)–MutL(α) sliding clamps that diffuse along the DNA to interact with the strand-discrimination signal (β-clamp/PCNA or MutH) (16). Other models include trapping of MutS(α) clamps near the mismatch by MutL(α) followed by DNA looping or, alternately, MutS(α)-induced polymerization of MutL(α) along the DNA to reach the strand-discrimination signal (6, 7, 15, 18, 22). Some degree of localization to the mismatch is suggested by in vitro studies of eukaryotic MMR proteins, indicating that although MutLα can introduce nicks across long stretches of DNA, they occur preferentially in the vicinity of the mismatch (9, 11, 12).In this study, we have used single molecule fluorescence to demonstrate that in the case of Thermus aquaticus (a MutH-free organism), MutL traps MutS at the mismatch after its ATP-induced activation but before its conversion into a sliding clamp. The resulting MutS–MutL mismatch complex typically contains more MutL than MutS, with one or two MutS dimers and up to four MutL dimers. MutS exists in a conformationally dynamic state within these complexes, which may be relevant for subsequent steps in MMR. In contrast to a mobile MutS–MutL complex, localization of MutS–MutL at the mismatch can restrict β-clamp/PCNA-activated MutL nicking to the vicinity of the mismatch, thereby enhancing MMR efficiency and limiting excessive excision and resynthesis that can destabilize the genome.     

Direct observation of DNA bending/unbending kinetics in complex with DNA-bending protein IHF

Regulation of gene expression involves formation of specific protein-DNA complexes in which the DNA is often bent or sharply kinked. Kinetics measurements of DNA bending when in complex with the protein are essential for understanding the molecular mechanism that leads to precise recognition of specific DNA-binding sites. Previous kinetics measurements on several DNA-bending proteins used stopped-flow techniques that have limited time resolution of few milliseconds. Here we use a nanosecond laser temperature-jump apparatus to probe, with submillisecond time resolution, the kinetics of bending/unbending of a DNA substrate bound to integration host factor (IHF), an architectural protein from Escherichia coli. The kinetics are monitored with time-resolved FRET, with the DNA substrates end-labeled with a FRET pair. The temperature-jump measurements, in combination with stopped-flow measurements, demonstrate that the binding of IHF to its cognate DNA site involves an intermediate state with straight or, possibly, partially bent DNA. The DNA bending rates range from approximately 2 ms(-1) at approximately 37 degrees C to approximately 40 ms(-1) at approximately 10 degrees C and correspond to an activation energy of approximately 14 +/- 3 kcal/mol. These rates and activation energy are similar to those of a single A:T base pair opening inside duplex DNA. Thus, our results suggest that spontaneous thermal disruption in base-paring, nucleated at an A:T site, may be sufficient to overcome the free energy barrier needed to partially bend/kink DNA before forming a tight complex with IHF.     

Mutation detection by mismatch binding protein, MutS, in amplified DNA: application to the cystic fibrosis gene.

An experimental strategy for detecting heterozygosity in genomic DNA has been developed based on preferential binding of Escherichia coli MutS protein to DNA molecules containing mismatched bases. The binding was detected by a gel mobility-shift assay. This approach was tested by using as a model the most commonly occurring mutations within the cystic fibrosis (CFTR) gene. Genomic DNA samples were amplified with 5'-end-labeled primers that bracket the site of the delta F508 3-bp deletion in exon 10 of the CFTR gene. The renatured PCR products from homozygotes produced homoduplexes; the PCR products from heterozygotes produced heteroduplexes and homoduplexes (1:1). MutS protein bound more strongly to heteroduplexes that correspond to heterozygous carriers of delta F508 and contain a CTT or a GAA loop in one of the strands than to homoduplexes corresponding to homozygotes. The ability of MutS protein to detect heteroduplexes in PCR-amplified DNA extended to fragments approximately 500 bp long. The method was also able to detect carriers of the point mutations in exon 11 of the CFTR gene by a preferential binding of MutS to single-base mismatches in PCR-amplified DNA.     

Mutation detection with MutH, MutL, and MutS mismatch repair proteins.

Escherichia coli methyl-directed mismatch repair is initiated by MutS-, MutL-, and ATP-dependent activation of MutH endonuclease, which cleaves at d(GATC) sites in the vicinity of a mismatch. This reaction provides an efficient method for detection of mismatches in heteroduplexes produced by hybridization of genetically distinct sequences after PCR amplification. Multiple examples of transition and transversion mutations, as well as one, two, and three nucleotide insertion/deletion mutants, have been detected in PCR heteroduplexes ranging in size from 400 bp to 2.5 kb. Background cleavage of homoduplexes is largely due to polymerase errors that occur during amplification, and the MutHLS reaction provides an estimate of the incidence of mutant sequences that arise during PCR.     

A rhodium(III) complex for high-affinity DNA base-pair mismatch recognition

A rhodium(III) complex, rac-[Rh(bpy)(2)phzi](3+) (bpy, 2,2'-bipyridine; phzi, benzo[a]phenazine-5,6-quinone diimine) has been designed as a sterically demanding intercalator targeted to destabilized mismatched sites in double-helical DNA. The complex is readily synthesized by condensation of the phenazine quinone with the corresponding diammine complex. Upon photoactivation, the complex promotes direct strand scission at single-base mismatch sites within the DNA duplex. As with the parent mismatch-specific reagent, [Rh(bpy)(2)(chrysi)](3+) [chrysene-5,6-quinone diimine (chrysi)], mismatch selectivity depends on the helix destabilization associated with mispairing. Unlike the parent chrysi complex, the phzi analogue binds and cleaves with high affinity and efficiency. The specific binding constants for CA, CC, and CT mismatches within a 31-mer oligonucleotide duplex are 0.3, 1, and 6 x 10(7) M(-1), respectively; site-specific photocleavage is evident at nanomolar concentrations. Moreover, the specificity, defined as the ratio in binding affinities for mispaired vs. well paired sites, is maintained. The increase in affinity is attributed to greater stability in the mismatched site associated with stacking by the heterocyclic aromatic ligand. The high-affinity complex is also applied in the differential cleavage of DNA obtained from cell lines deficient in mismatch repair vs. those proficient in mismatch repair. Agreement is found between photocleavage by the mismatch-specific probes and deficiency in mismatch repair. This mismatch-specific targeting, therefore, offers a potential strategy for new chemotherapeutic design.     

DNA bending by hexamethylene-tethered ammonium ions.

DNA is bent when complexed with certain proteins. We are exploring the hypothesis that asymmetric neutralization of phosphate charges will cause the DNA double helix to collapse toward the neutralized face. We have previously shown that DNA spontaneously bends toward one face of the double helix when it is partially substituted with neutral methylphosphonate linkages. We have now synthesized DNA duplexes in which cations are tethered by hexamethylene chains near specific phosphates. Electrophoretic phasing experiments demonstrate that tethering six ammonium ions on one helical face causes DNA to bend by approximately 5 degrees toward that face, in qualitative agreement with predictions. Ion pairing between tethered cations and DNA phosphates provides a new model for simulating the electrostatic consequences of phosphate neutralization by proteins.     

A mismatch recognition defect in colon carcinoma confers DNA microsatellite instability and a mutator phenotype.

We have analyzed spontaneous mutations in the adenine phosphoribosyltransferase gene of Chinese hamster clone B cells that exhibit a mutator phenotype because of defective mismatch binding. The mutator phenotype conferred increases in a limited number of mutational classes. The rates of transitions and most transversions were not significantly increased. The rates of A to T transversions and -2 frameshifts were strikingly elevated. These mutations were in repeated elements and 5 of 9 of the frameshifts were dinucleotide deletions in DNA sequences resembling microsatellites. The mismatch binding protein that is defective in the mutator line is a G-T mismatch recognition factor. Band-shift analysis indicated that the preferred substrate for the mismatch recognition protein is duplex DNA containing an extrahelical mono- or dinucleotide within repeated sequences. In agreement with a role in preventing minus frameshifts, a defective binding protein conferred an instability in clone B microsatellite DNA. A mismatch binding defect was also detected in Lo Vo, a human colorectal carcinoma cell line. Extracts of clone B or a second mismatch binding-deficient line, Raji-F12, did not complement Lo Vo extracts, indicating that these lines share a common defect. Our data provide a mechanistic explanation for the relation between defective mismatch recognition and the microsatellite instability of human colon cancer.     

Conformational trapping of mismatch recognition complex MSH2/MSH3 on repair-resistant DNA loops

Insertion and deletion of small heteroduplex loops are common mutations in DNA, but why some loops are prone to mutation and others are efficiently repaired is unknown. Here we report that the mismatch recognition complex, MSH2/MSH3, discriminates between a repair-competent and a repair-resistant loop by sensing the conformational dynamics of their junctions. MSH2/MSH3 binds, bends, and dissociates from repair-competent loops to signal downstream repair. Repair-resistant Cytosine-Adenine-Guanine (CAG) loops adopt a unique DNA junction that traps nucleotide-bound MSH2/MSH3, and inhibits its dissociation from the DNA. We envision that junction dynamics is an active participant and a conformational regulator of repair signaling, and governs whether a loop is removed by MSH2/MSH3 or escapes to become a precursor for mutation.     

Chromatin remodeling by DNA bending, not twisting

Single-stranded regions (gaps) in nucleosomal DNA interfere with action of the RSC chromatin-remodeling complex, monitored by exposure of restriction endonuclease cutting sites. Single-strand breaks (nicks) in the DNA, by contrast, have no effect. Gaps on one side of the cutting site are inhibitory, but gaps on the other side are not. A gap >100 bp from the cutting site is as effective as a gap <20 bp from the site. These findings suggest a remodeling process involving bending, but not twisting, of the DNA and further point to the propagation of a bent region (loop or bulge) from one end of the nucleosome to the other.     

O6-methylguanine-induced cell death involves exonuclease 1 as well as DNA mismatch recognition in vivo

Alkylation-induced O⁶-methylguanine (O⁶MeG) DNA lesions can be mutagenic or cytotoxic if unrepaired by the O⁶MeG-DNA methyltransferase (Mgmt) protein. O⁶MeG pairs with T during DNA replication, and if the O⁶MeG:T mismatch persists, a G:C to A:T transition mutation is fixed at the next replication cycle. O⁶MeG:T mismatch detection by MutSα and MutLα leads to apoptotic cell death, but the mechanism by which this occurs has been elusive. To explore how mismatch repair mediates O⁶MeG-dependent apoptosis, we used an Mgmt-null mouse model combined with either the Msh6-null mutant (defective in mismatch recognition) or the Exo1-null mutant (impaired in the excision step of mismatch repair). Mouse embryonic fibroblasts and bone marrow cells derived from Mgmt-null mice were much more alkylation-sensitive than wild type, as expected. However, ablation of either Msh6 or Exo1 function rendered these Mgmt-null cells just as resistant to alkylation-induced cytotoxicity as wild-type cells. Rapidly proliferating tissues in Mgmt-null mice (bone marrow, thymus, and spleen) are extremely sensitive to apoptosis induced by O⁶MeG-producing agents. Here, we show that ablation of either Msh6 or Exo1 function in the Mgmt-null mouse renders these rapidly proliferating tissues alkylation-resistant. However, whereas the Msh6 defect confers total alkylation resistance, the Exo1 defect leads to a variable tissue-specific alkylation resistance phenotype. Our results indicate that Exo1 plays an important role in the induction of apoptosis by unrepaired O⁶MeGs.     

Cue utilization and encoding specificity in picture recognition by older adults

According to the encoding specificity principle, memory is best when encoding and retrieval conditions are compatible. Some researchers have suggested that older adults encode information in a general fashion and are less sensitive to the specific contextual aspects of a memory situation due to limited processing resources. We investigated the hypothesis that age interacts with encoding specificity. Young and old adults studied target pictures in the presence or absence of pictorial cues factorially varied at encoding and retrieval. If the older adults used the specific cuing information differently from the younger adults, age should have interacted with the encoding and retrieval variables. The results provided no evidence for such an interaction and indicated that both ages showed evidence of encoding specificity. To investigate the role of processing resources in encoding specificity, old and young adults also studied the pictures while simultaneously performing a digit-monitoring task. The divided-attention manipulation also did not interact with age, as both young and old adults showed encoding specificity effects of comparable magnitude in both control and divided-attention conditions.     

DNA damage-site recognition by lysine conjugates

Simple lysine conjugates are capable of selective DNA damage at sites approximating a variety of naturally occurring DNA-damage patterns. This process transforms single-strand DNA cleavage into double-strand cleavage with a potential impact on gene and cancer therapy or on the design of DNA constructs that require disassembly at a specific location. This study constitutes an example of DNA damage site recognition by molecules that are two orders of magnitude smaller than DNA-processing enzymes and presents a strategy for site-selective cleavage of single-strand nucleotides, which is based on their annealing with two shorter counterstrands designed to recreate the above duplex damage site.     

Protein-induced bending and DNA cyclization.

We have applied T4 ligase-mediated DNA cyclization kinetics to protein-induced bending in DNA. The presence and direction of a static bend can be inferred from J factors for cyclization of 150- to 160-base-pair minicircles, which include a catabolite activator protein binding site phased against a sequence-directed bend. We demonstrate a quasi-thermodynamic linkage between cyclization and protein binding; we find that properly phased DNAs bind catabolite activator protein approximately 200-fold more tightly as circles than as linear molecules. The results unambiguously distinguish DNA bends from isotropically flexible sites and can explain cooperative binding by proteins that need not contact each other.     

Nucleotide specificity in DNA scission by neocarzinostatin.

Using the DNA sequencing technique of Maxam and Gilbert, we show that the protein antibiotic neocarzinostatin, cleaves double-stranded phiX174 DNA restriction fragments almost exclusively at deoxythymidylic and deoxyadenylic acid residues in a reaction requiring 2-mercaptoethanol. Overall, deoxythymidylic acid residues are attacked much more frequently than are deoxyadenylic acid residues, although there is variability in the attack rate for both nucleotides at different locations in the DNA molecule. While all deoxythymidylic acid residues are sites of scission by neocarzinostatin, not all deoxyadenylic acid residues are cleavage sites. There appears to be no clear-cut nucleotide sequence specificity in determining cleavage frequency. Single-stranded DNA is a very poor substrate for neocarzinostatin-induced scission; with one single-stranded DNA fragment, cleavage occurs at a position that is not attacked in double-stranded DNA. The possible significance for its biological activity of a drug that can attack both members of a DNA base pair is discussed.     

Molecular basis of DNA sequence recognition by the catabolite gene activator protein: detailed inferences from three mutations that alter DNA sequence specificity.

Previously, we reported that substitution of Glu-181 of the catabolite gene activator protein (CAP) by lysine, leucine, or valine results in a protein that has specificity for A X T base pairs at positions 7 and 16 of the DNA recognition site, rather than G X C base pairs as is the case with the wild-type CAP. In this paper, we deduce from these genetic data both (i) the specific chemical interactions by which amino acid side chains at position 181 interact with base pairs 7 and 16 and (ii) the precise alignment between the structures of the CAP and DNA in the intermolecular CAP-DNA complex. Our analysis supports the idea that the two symmetry-related F alpha-helices of the CAP dimer interact with successive major grooves of right-handed B-type DNA [Pabo, C. & Lewis, M. (1982) Nature (London) 298, 443-447; and Steitz, T., Weber, I. & Matthew, J. (1983) Cold Spring Harbor Symp. Quant. Biol. 47, 419-426].     

Stepwise binding and bending of DNA by Escherichia coli integration host factor

Integration host factor (IHF) is a prokaryotic protein required for the integration of lambda phage DNA into its host genome. An x-ray crystal structure of the complex shows that IHF binds to the minor groove of DNA and bends the double helix by 160 degrees [Rice PA, Yang S, Mizuuchi K, Nash HA (1996) Cell 87:1295-1306]. We sought to dissect the complex formation process into its component binding and bending reaction steps, using stopped-flow fluorimetry to observe changes in resonance energy transfer between DNA-bound dyes, which in turn reflect distance changes upon bending. Different DNA substrates that are likely to increase or decrease the DNA bending rate were studied, including one with a nick in a critical kink position, and a substrate with longer DNA ends to increase hydrodynamic friction during bending. Kinetic experiments were carried out under pseudofirst-order conditions, in which the protein concentration is in substantial excess over DNA. At lower concentrations, the reaction rate rises linearly with protein concentration, implying rate limitation by the bimolecular reaction step. At high concentrations the rate reaches a plateau value, which strongly depends on temperature and the nature of the DNA substrate. We ascribe this reaction limit to the DNA bending rate and propose that complex formation is sequential at high concentration: IHF binds rapidly to DNA, followed by slower DNA bending. Our observations on the bending step kinetics are in agreement with results using the temperature-jump kinetic method.     

Static and statistical bending of DNA evaluated by Monte Carlo simulations.

To investigate the influence of thermal fluctuations on DNA curvature the Metropolis procedure at 300 K was applied to B-DNA decamers containing A5.T5 and A4.T4 blocks. Monte Carlo simulations have confirmed the DNA bending anisotropy: B-DNA bends most easily in a groove direction (roll). The A5.T5 block is more rigid than the other sequences; the pyrimidine-purine dimers are found to be the most flexible. For A5TCTCT, A5CTCTC, and A5GAGAG, the average bend angle per decamer is 20-25 degrees in a direction toward the minor groove in the center of the A5.T5 tract, which is consistent with both the "junction" and "wedge AA" models. However, in A5T5, A4T4CG, and T4A4GC, bending is directed into the grooves at the 5' and 3' ends of purine tracts. Thus, directionality of bending caused by An.Tn blocks strongly depends on their neighboring sequences. These calculations demonstrate that the sequence-dependent variation of the minor-groove width mimics the observed hydroxyl radical cleavage pattern. To estimate the effect of fluctuations on the overall shape of curved DNA fragments, longer pieces of DNA (up to 200 base pairs) were generated. For sequences with strong curvature (A5X5 and A4T4CG), the static model and Monte Carlo ensemble give similar results but, for moderately and slightly curved sequences (A5T5 or T4A4GC), the static model predicts a much smaller degree of bending than does the statistical representation. Considering fluctuations is important for quantitative interpretation of the gel electrophoresis measurements of DNA curvature, where both the static and statistical bends are operative.