› content › 286 › 19 › 17326.full.pdf · chemicaltrappingofthedynamicmuts-mutlcomplex...

Chemical Trapping of the Dynamic MutS-MutL ComplexFormed in DNA Mismatch Repair in Escherichia coli*□S

Received for publication, September 29, 2010, and in revised form, February 24, 2011 Published, JBC Papers in Press, March 15, 2011, DOI 10.1074/jbc.M110.187641

Ines Winkler‡, Andreas D. Marx‡, Damien Lariviere§, Roger J. Heinze‡, Michele Cristovao‡¶, Annet Reumer�,Ute Curth**, Titia K. Sixma�, and Peter Friedhoff‡1

From the ‡Institute for Biochemistry, FB 08, Justus Liebig University, D-35392 Giessen, Germany, §Fourmentin-Guilbert ScientificFoundation, Avenue du Pave Neuf, 93160 Noisy-Le-Grand, France, �Division of Biochemistry and Center for Biomedical Genetics,Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands, **Institute for Biophysical Chemistry,Hannover Medical School, D-30625 Hannover, Germany, and ¶Department of Cell Biology and Genetics, Erasmus Medical Center,P. O. Box 2040, 3000 CA Rotterdam, The Netherlands

The ternary complex comprising MutS, MutL, and DNA is akey intermediate in DNA mismatch repair. We used chemicalcross-linking and fluorescence resonance energy transfer(FRET) to study the interaction betweenMutS andMutL and toshed light onto the structure of this complex. Via chemicalcross-linking, we could stabilize this dynamic complex andidentify the structural features of key events in DNA mismatchrepair. We could show that in the complex between MutS andMutL the mismatch-binding and connector domains of MutSare in proximity to theN-terminalATPase domainofMutL.TheDNA- and nucleotide-dependent complex formation could bemonitored by FRET using single cysteine variants labeled in theconnector domain ofMutS and the transducer domain ofMutL,respectively. In addition, we could trap MutS after an ATP-in-duced conformational change by an intramolecular cross-linkbetween Cys-93 of the mismatch-binding domain and Cys-239of the connector domain.

The DNA mismatch repair system (MMR)2 plays an impor-tant role in maintaining genomic stability and avoidance ofmutations in the DNA sequence (1). Major landmark studies inMMR have been the reconstitution of the system from purifiedproteins both for Escherichia coli and humans (2–4). The keyMMR proteins MutS and MutL are present in bacteria andeukaryotes, and the initial steps of MMR are conserved (5).Whereas bacterial MutS and MutL proteins are homodimericproteins, the eukaryotic homologues are heterodimeric pro-teins; e.g. MutS� is composed of MutS homologues 2 and 6(MSH2/MSH6), andMutL� consists ofMLH1 and PMS2 (6, 7).

Bacterial MutS exists in a dimer-tetramer equilibrium, but thedimeric form of the protein is sufficient for DNA mismatchrepair (8, 9). MutS is composed of seven domains (mismatch-binding, connector, core, lever, clamp, ATPase, and dimeriza-tion/tetramerization domains) (see Fig. 1) (9–12). MutL hasbeen dissected into a 40-kDa N-terminal half (LN40) and a20-kDaC-terminal dimerization domain (LC20) (13–16). LN40comprises the ATPase domain and transducer domain and isconnected by a non-conserved linker to the LC20 domain(13, 14).In crystal structures, an asymmetric complex of MutS encir-

cling the heteroduplex DNA was observed. In this asymmetriccomplex, only the conserved Phe-X-Glu-motif of the mis-match-binding domain of one subunit is in direct contact withthe mismatch, resulting in a 45–60° kink at the mismatch (10,11, 17–19). In E. coli, the mismatched base is stacked ontoPhe-36 and hydrogen-bonded to Glu-38, and stacking ofPhe-36 between the bases is important for mismatch recogni-tion and prevents MutS from sliding on the DNA (20). A hall-mark of themismatch recognition process byMutS is the ATP-induced transition from a stationary clamp bound to themismatched DNA to a long livedmobile sliding clamp (21–23).For the formation of the sliding clamp, conformational changesinvolving movement of the mismatch-binding domain havebeen proposed (24). Computational studies using normalmodeanalysis and molecular dynamics have supported these models(25, 26).ATP andMg2� binding was shown to be crucial not only for

the formation of a sliding clamp but also for the formation of adynamic ternary complex comprising DNA-MutS-MutL (27,28). This ternary complex coordinates all subsequent steps inDNA repair (e.g. strand discrimination and DNA unwinding)and is also involved in signaling the DNAmismatch/damage toother cellular responses (29). Several lines of evidence suggestthat MutS undergoes substantial conformational changes dur-ing this transition; however, until recently few structural detailswere known (30, 31).A recent studywith yeastMutS� suggested that ternary com-

plex formation and sliding clamp formation are distinct stepsinvolving different nucleotide states of MutS. Moreover, it wassuggested that the ternary complex precedes sliding clamp for-mation (30). Attempts to get high resolution structural infor-mation on the ternary complex or the sliding clamp, e.g. by

* This work was supported by European Community Seventh FrameworkProgram FP7/2007-2013 under Grant HEALTH-F4-2008-223545 (to P. F.,I. W., and R. J. H.), Deutsche Forschungsgemeinschaft Grant GRK 1384“Enzymes and Multienzyme Complexes Acting on Nucleic Acids,” and Rus-sian Foundation for Basic Research Grant RFBR-DFG 08-04-91973 (to P. F.and A. D. M.).

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. S1–S4, Tables S1 and S2, and additional experimentalprocedures.

1 To whom correspondence should be addressed: Inst. for Biochemistry, Hei-nrich-Buff-Ring 58, D-35392 Giessen, Germany. Fax: 49-641-99-35407;E-mail: [email protected].

2 The abbreviations used are: MMR, mismatch repair system; AMPPNP, 5�-ad-enylyl-�,�-imidodiphosphate; BM[PEO]4, 1,11-bismaleimidotetraethyleneglycol; ATP�S, adenosine 5�-O-(thiotriphosphate).

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 19, pp. 17326 –17337, May 13, 2011© 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

17326 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 19 • MAY 13, 2011

by guest on March 2, 2020

http://ww

w.jbc.org/

Dow

nloaded from


http://ww

w.jbc.org/

Dow

nloaded from


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/cgi/content/full/M110.187641/DC1


http://www.jbc.org/

http://www.jbc.org/

http://www.jbc.org/

x-ray crystallography, have been hampered by the dynamicnature of the ternary complex, the mobility, and lack of DNAsequence specificity of the MutS sliding clamp. Mutationalanalyses and hydrogen/deuterium exchange mass spectrome-try analyses suggested that the connector domain ofMutS con-tains residues critical for the interaction with MutL (32).Here, we used Cys-mediated cross-linking to shed light onto

these two key events in DNAmismatch repair. To this end, wegenerated a set of cysteine variants of MutS and MutL andtested the proteins for cross-linking using the highly specificand reactivemaleimide cross-linking chemistry. In addition, weused site-specifically labeled MutS and MutL single cysteinevariants to monitor the ternary complex formation in solutionusing fluorescence resonance energy transfer (FRET). Based onour results and data from the literature, we provide a model ofthe MutS-MutL-MutH complex and its implication for theMMR process.

EXPERIMENTAL PROCEDURES

DNA Substrates—Linear DNA substrates (484 bp with orwithout a GTmismatch) were generated from two 484-bp PCRproducts amplified by Pfu DNA polymerase with a 5�-phos-phate at the top or the bottom strand, respectively, using plas-mids pET-15b-XhoI (with a GC base pair) and/or pET-15b-HindIII (with an AT base pair), respectively, with primers(BBseq A302/pA302, 5�-ATC TTC CCC ATC GGT GATGTC-3�; and BBseq B111/p111, 5�-TCA TCC TCG GCA CCGTCA C-3�) essentially as described previously (33). The phos-phorylated strands were digested by �-exonuclease, and the topstrand (containing a G at position 385) was annealed to thebottom strand (containing a C or a T at position 385, respec-tively) to give either a 484-bp homoduplex (GC484) or hetero-duplex (GT484), respectively. A 42-bp GT heteroduplex(GT42) corresponding to positions 365–406 of GT484 wasgenerated by annealing the two oligonucleotides TAT TAATTT CGC GGG CTC GAG AGC TTC ATC CTC TAC GCCGGA and TCC GGC GTA GAG GAT GAA GCT TTC GAGCCC GCG AAA TTA ATA (the underlined bases indicate theposition of the GTmismatch). The generation of circular DNAsubstrates (described in detail in the supplemental informa-tion) containing a single GT mismatch within a hemimethyl-ated Dcm site at position 169 and a hemimethylated GATC siteat position 356 (3315 bp with GT mismatch) was performedessentially as published before (34).Site-directedMutagenesis—Plasmids encoding the gene for a

cysteine-free MutS (MutS-CF) and dimeric MutS-CF/D835Rhave been described before (8). Single cysteine MutS variants(Table 1 and supplemental Table S1) were generated by site-directed mutagenesis using a modification of the QuikChangeprotocol (35, 36). E. coliXL1-blueMRF� cells were transformedwith the full-length PCR product, and marker-positive cloneswere inoculated and grownovernight in LBmediumwith ampi-cillin. PlasmidDNAwas isolated usingQIAprep SpinMiniprep(Qiagen) or Wizard� Plus Miniprep (Promega), and the entiremutS gene was sequenced to confirm the mutation.Protein Expression and Purification—Recombinant His6-

tagged proteins (MutS and MutL) and variants thereof wereexpressed and purified by nickel-nitrilotriacetic acid chroma-

tography and size exclusion chromatography essentially asdescribed elsewhere (8, 36–38). MutS andMutL proteins werestored in 10 mMHEPES/KOH (pH 7.9), 200 mM KCl, and 1 mM

EDTA (for MutS proteins 10% glycerol was added). MutS andMutL were snap frozen in liquid nitrogen and stored at�80 °C.Protein concentrations were determined from UV absorbancespectra using the theoretical molar extinction coefficients(MutS, 68,000 M�1 cm�1; or MutL, 54,800 M�1 cm�1) and aregiven in monomeric equivalents (39).Mismatch-provoked Activation of MutH Endonuclease—

Variants of MutS andMutL were tested for their ability to acti-vate the MutH endonuclease using a covalently closed circularheteroduplex DNA substrate (3315 bp) containing a GT mis-match and a hemimethylated GATC site separated by 186 bp.DNA (15 nM) was incubated with 200 nMMutH, 400 nMMutL,and 400 nM MutS in 100 �l of 10 mM Tris-HCl (pH 7.9), 5 mM

MgCl2, 1mMATP, and 125mMKCl buffer at 37 °C.Aliquots (10�l) were taken after different time points (0, 10, 20, 30, 60, 120,300, and 600 s), and the reactions were stopped with 2 �l of 250mM EDTA, 25% sucrose, 1.25% SDS, 0.1% bromphenol blue,and 1 �l of 10 units/�l proteinase K. Samples were subjected togel electrophoresis (1% agarose in Tris-phosphate-EDTA con-taining 0.5 �g/ml ethidium bromide). MutH activity was mon-itored by the conversion of covalently closed circular DNA tonicked DNA (open circular). The intensity of the ethidium bro-mide-stained DNA bands was quantified using TotalLab v2.01software and analyzed with Origin8.5 software. A single expo-nential functional was used to fit the time course to yield appar-ent first order rate constants (kobs).Chemical Cross-linking—The homobifunctional maleimide

cross-linkers of varying length, 1,11-bismaleimidotetraethyl-ene glycol (BM[PEO]4) and bismaleimidoethane, and themeth-anethiosulfonate cross-linker were obtained from Pierce andToronto Research Chemicals, respectively. Stock solutions of10 mM were made in water or DMSO. MutL (10 �M) variantswere incubated with 5 mM nucleotide (ADP, ATP, or AMP-PNP) andMutS (0.57�M) variants with homo- or heteroduplexDNA on ice for 25 min in buffer HK (20 mM HEPES/KOH (pH7.5), 5 mMMgCl2, and 0.01mM EDTA) and 125 or 150mMKCl.MutS and MutL were mixed with final concentrations as indi-cated. After 10-min incubation at room temperature followedby 2 min at 37 °C, 50 �M cross-linker was added to the samples(at least 50-fold molar excess over thiol groups) and incubatedfor 1–2 min at 37 °C. The reaction was quenched by addingDTT (50mM final concentration formaleimide cross-linker) orN-ethylmaleimide (50 mM final concentration for methaneth-iosulfonate cross-linker). Samples were subjected to 6% SDS-PAGE and stainedwith colloidal Coomassie Blue (AppliChem).Gels were analyzed with a video documentation system (Bio-Rad). The intensity of the stained protein bands was quantifiedusing TotalLab v2.01 software. Cross-linking yields were calcu-lated as

% S � L �

I�S � L�

Mr,S � Mr,L

I�S�

Mr,S�

I�S � L�

Mr,S � Mr,L�

I�S � S�

Mr,S

(Eq. 1)

Chemical Trapping of MutS-MutL Complex

MAY 13, 2011 • VOLUME 286 • NUMBER 19 JOURNAL OF BIOLOGICAL CHEMISTRY 17327


http://ww

w.jbc.org/

Dow

nloaded from




http://www.jbc.org/

where I (S x L) is the intensity of the MutS-MutL band, I (S x S)is the intensity of theMutS-MutS band, and I (S) is the intensityof the MutS band.Mr,S � 97,000 andMr,L � 70,000, i.e. theMr(molecular weight) of MutS and MutL, respectively.Labeling Proteins with Fluorophores—MutS and MutL vari-

ants were labeled with Alexa Fluor 488 or Alexa Fluor 594 in a1:3 or 1:4 molar ratio, respectively. After 30-min incubation onice, the samples were twice purified over Zeba Desalt Spin Col-umns (Pierce, Thermo Scientific) followed by themeasurementof the concentration using absorbance spectroscopy,

cprot �A280 Amax � CF

prot(Eq. 2)

whereA280 is the absorbance at 280 nm,Amax is the absorbanceof the Alexa Fluor 594 or Alexa Fluor 488, CF is the correctionfactor for the used dye (CFA488 � 0.12 and CFA594 � 0.57), andprot is the theoretical molar extinction coefficients of therespective protein at 280 nm. The degree of labeling (DOL) wasdetermined using the equation

DOL �Amax � prot

� A280 Amax � CF� � max(Eq. 3)

where max is the molar extinction coefficients of Alexa Fluor488 (71,000 M�1 cm�1) or Alexa Fluor 594 (73,000 M�1 cm�1),respectively.Analytical Ultracentrifugation—Sedimentation velocity ex-

periments were performed in a BeckmanCoulter ProteomeLabXL-I analytical ultracentrifuge equipped with a fluorescencedetection system (Aviv Biomedical) using an An50Ti rotor at20 °C and speeds from22,000 to 33,000 rpm.The concentrationprofiles were measured using the analytical ultracentrifugeequipped with a fluorescence detection system with an excita-tion wavelength of 488 nm, and emission was detected througha pair of long pass (�505-nm) dichroic filters. Programming ofthe centrifuge and data recording were performed using theAOS software (Aviv Biomedical) on a computer attached to thecentrifuge. Special cell housings (Nanolytics) were used thatallow the placement of standard 3-mm double sector center-pieces directly beneath the upper window of the cell. The cellswere filled with 100 �l of sample. The experiments were per-formed in buffer HKT (20 mM HEPES (pH 7.5), 125 mM KCl, 5mM MgCl2, 0.01 mM EDTA, and 0.05% (v/v) Tween 20 to pre-vent protein adsorption to surfaces). For MutS-449C andMutL-297C, the Alexa Fluor 488-labeled proteins were exam-ined at concentrations between 100 nM and 1 �M. The experi-ments for MutS-246C-Alexa Fluor 488 were done at a concen-tration of 100 nM. To test the influence of bound nucleotide onthe sedimentation behavior, the experiments were performedin the presence of 1 mM ADP, ATP, or AMPPNP, respectively.To check whether the fluorescently labeled MutS variants arestill able to bind to heteroduplex DNA, a 100 nM concentrationof the respective protein was incubated in the presence of 1mM

ADP with 500 nM GT42 and subjected to sedimentation veloc-ity analysis. The measured concentration profiles were evalu-ated using the program package SEDFIT (40), which providesa model for diffusion-corrected differential sedimentationcoefficient distributions (c(s) distributions). For hydrodynamic

analyses, measured s values were corrected to s20,w using thepartial specific volumes calculated from amino acid composi-tion (41). Because the partial specific volume of complexes ofdifferent macromolecules with unknown composition cannotbe calculated, such a correction could not be performed, anduncorrected sedimentation coefficients (sexp) are given in thesecases.Fluorescence Spectroscopy—Fluorescence emission spectra

were obtained using a FluoroMax-4 (HORIBA JobinYvon)withexcitation wavelength �ex at 470 or 575 nm and band width setto 1.8 nm. Emission spectra were recorded at 20 °C in a totalvolume of 100�l in bufferHKT. Spectrawere normalized to themaximum of fluorescence intensity at 515 nm.Protein-Protein Docking—Docking runs were performed

with the Hex software to automatically get thousands of con-figurations (42). In a first docking run, we used the mismatch-binding and connector domains of MutS (Protein Data Bankcode 1e3m, chain B, residues 2–266) and the N-terminaldomains ofMutL (Protein Data Bank code 1b63, chain A and B,residues 1–331), and in a second docking run, the N-terminaldomains of MutL and MutH (Protein Data Bank code 2azo,chain B) (37) were used. Docking solutions were used to gener-ate structural models of the full-length MutS with MutL andMutH. These models were filtered using the experimentallyderived distance constraints (this work and Ref. 37). For detailsof the procedure, see the supplemental information.

RESULTS

Cys-93 in Mismatch-binding Domain of MutS Cross-links toCys-131 in ATPase Domain of MutL—In a previous report, wedemonstrated that the single cysteine MutL variant (MutL-131C) modified with benzophenone at Cys-131 in the ATPasedomain could be photo-cross-linked to MutS (43). In addition,using the thiol-specific bismaleimide reagent BM[PEO]4 (Fig.1A), we were able to obtain a MutS-MutL cross-link in thepresence of a GTmismatch DNA and ATP (43). This indicatedthat one or more of the six endogenous cysteine residues ofMutS are in proximity to the N-terminal domain of MutL (Fig.1B). Here we investigated the cross-link formation betweenMutS and MutL-131C in more detail. In the presence of ATP,the formation of a cross-linkwas dependent on the length of theDNA and was observed with a long 484-bp GT-containingDNA (Fig. 1C, lane 2) butwas almost absentwhen theDNAwasonly 42 bp long (Fig. 1C, lane 5). Moreover, the cross-link for-mation was mismatch-dependent as we only observed theMutS-MutL cross-link in the presence of heteroduplex (GT)DNA but not with the corresponding homoduplex (GC) DNA(supplemental Fig. S1). To test whether ATP binding or ATPhydrolysis is required to form the cross-link, we replaced ATPby ADP or the non-hydrolyzable analogue AMPPNP. It hasbeen noticed before that MutS is not able to recognize mis-matches when ATP is bound but cannot be hydrolyzed, e.g. inthe absence of Mg2� or by using non-hydrolyzable ATP ana-logueATP�S orAMPPNP (44–46). As expected from theDNAdependence of the cross-link formation, no MutS-MutL cross-link was observed when MutS was incubated with AMPPNPprior to the addition of DNA and MutL (data not shown). Incontrast, when MutS was bound to DNA prior to the addition




http://ww

w.jbc.org/

Dow

nloaded from



http://www.jbc.org/

of MutL and AMPPNP, MutS-MutL cross-link formation wasas efficient as with ATP, indicating that ATP binding but nothydrolysis is required for cross-link formation (Fig. 1C, lanes 2and 4). No cross-link formationwas observed in the presence ofADP (Fig. 1C, lane 3). Moreover, using MutS-E694A, which isdefective in ATP hydrolysis (47, 48), cross-link formation wasobserved only whenMutS was first bound to DNA followed bythe addition of ATP, MutL, and the cross-linker (data notshown).To identify the cysteine residue involved in the chemical

cross-linking reaction, we tested variants in which one or more

of the endogenous cysteine residues were replaced by otheramino acid residues (supplemental Table S2). The six non-con-served cysteine residues in MutS are located in the mismatch-binding domain (Cys-93), the connector domain (Cys-235 andCys-239), the core domain (Cys-297), and the ATPase domain(Cys-569 and Cys-711) (Fig. 1B). In the crystal structure ofMutS in complex withmismatchedDNA and boundADP, onlythree of these cysteine residues (235, 297, and 569) are solvent-exposed (49). By replacing all six cysteine residues, we gener-ated a cysteine-free variant of MutS that is fully functional invitro and in vivo, indicating that none of the cysteine residuesare essential for MutS function (8). None of the three solvent-exposed cysteine residues (235, 297, and 569) were essential forchemical cross-link formation withMutL because replacementof these residues with serine did not abolish cross-link forma-tion with MutL-131C (supplemental Table S2). Moreover, thesingle cysteine MutS variants MutS-297C and MutS-569C didnot form a cross-link to MutL-131C (supplemental Table S2).In contrast, all MutS variants still harboring Cys-93, which islocated in themismatch-binding domain, were able to form thechemical cross-link with MutL-131C. Therefore, we generateda single cysteine variant, MutS-93C, containing only Cys-93and compared the MutL cross-linking reactions with that withwild type MutS (Fig. 1C). MutS-93C displayed the same nucle-otide and DNA dependence as wild type MutS; i.e. efficientcross-link formationwas observed only in the presence of DNAand ATP or AMPPNP but not ADP (Fig. 1C, lanes 8–10).Finally, efficient cross-linking between MutS-93C and MutL-131C was even observed using the shorter cross-linker bisma-leimidoethane, which has a calculated sulfur to sulfur distancerange of about 6.3–10.5 Å (data not shown) (50). In summary,our data revealed that the chemical cross-linking betweenCys-93 of the mismatch-binding domain of MutS and Cys-131of the N-terminal domain of MutL occurs only under condi-tions that are required for ternary complex formation asobserved in other biochemical assays (51). Given the length ofthe cross-linker used, our cross-linking data indicate that in theternary complex the mismatch-binding domain of MutS is inclose proximity (less than 10Å) to theATPase domain ofMutL.Cys-246 of Connector Forms Cross-link to Cys-131 of MutL—

To obtain additional distance restraints and to get furtherinsights into structure of the ternary complex, we generated aseries of single cysteine MutS variants (supplemental Fig. S2).The positions were chosen on the basis of two criteria. First, theresidues to be replaced should bemaximally solvent-exposed toallow optimal reaction with the cross-linker. Second, the resi-dues should not be conserved to minimize negative effects ofthe amino acid exchange.A list of the variants tested is shown inTable 1 (see supplemental Fig. S2 for the position of cysteineresidues inMutS tested in the present study). Indeed, almost allplasmid-borneMutS variants were able to complement amutSmutator phenotype in vivo, indicating that the proteins werefunctional and still able to interact with MutL. Next we testedthe purified MutS variants for their ability to form a cross-linkwith MutL-131C. Interestingly, only six variants with a singlecysteine either in the mismatch-binding domain (MutS-8C,MutS-78, MutS-93C, and MutS-103), the connector domain(MutS-246C), or the lever domain (MutS-427C) resulted in effi-

FIGURE 1. Cys-93 in mismatch-binding domain of MutS cross-links to Cys-131 in ATPase domain of MutL. A, structure and range of the thiol-specificchemical cross-linker BM[PEO]4 (50). B, crystal structure of MutS (residues1– 800) bound to GT DNA (Protein Data Bank code 1e3m (10)). The C-terminaldimerization/tetramerization domain (residues 801– 853) is missing. SubunitA is denoted in gray. Domains of subunit B are colored as follows: mismatch-binding domain (residues 1–115), blue; connector domain (residues 116 –266), cyan; core domain and levers (residues 267– 443 and 504 –567), red;clamp domain (residues 444 –503), orange; ATPase and helix-turn-helixdomain (residues 568 – 800), green. Cysteine residues are shown as yellowspheres. Residues that have been shown to display reduced solvent accessi-bility upon MutL binding are indicated as cyan (residues 205–213) and green(residues 678 – 688) spheres (32). C, nucleotide and DNA dependence of thecross-linking reaction between MutS and MutL. MutS (wild type or singlecysteine MutS-93C; 400 nM) was incubated with single cysteine MutL-131C(1000 nM) in the presence or absence of a GT heteroduplex DNA (100 nM

GT484 or 500 nM GT42), the indicated nucleotide (1 mM), and the chemicalcross-linker BM[PEO]4 (50 �M) for 2 min at 37 °C. Reaction mixtures were ana-lyzed by SDS-PAGE (6%) and colloidal Coomassie Blue staining. Note that theelectrophoretic mobilities of the MutS-MutL cross-link (S x L) (verified afterin-gel trypsin digestion and mass spectrometry) observed with MutS wildtype and MutS-93C are identical. Cross-link formation of MutL-MutL (L x L) hasbeen described before (37). Efficient cross-link formation of MutS and MutLwas only observed in the presence of ATP (lanes 2, 5, and 8) or AMPPNP (lanes4, 7, and 10) and long GT DNA. It was not observed in the presence of ADP(lanes 3, 6, and 9), with short DNA (lanes 5–7), or in the absence of DNA (lane11). No cross-link formation was observed when the cross-linker was omittedor when GT DNA was replaced by a corresponding homoduplex control DNA(supplemental Fig. S1).




http://ww

w.jbc.org/

Dow

nloaded from







http://www.jbc.org/

cient (�25%) cross-link formation in the presence of ATP(Table 1) (32). Next we investigated the kinetics of cross-linkformation for two variants (MutS-93C and MutS-246C) withMutL-131C in greater detail. Interestingly, kinetics of the cross-linking reaction revealed that the formation of theMutS-MutLcross-link was much faster (within 10 s) withMutS-246C com-paredwithMutS-93C (Fig. 2,A–C). AnotherMutS variant witha single cysteine in the connector domain (MutS-162C) wastested further for cross-linking to anotherMutL variant (MutL-135C). For this variant, cross-linking was not observed withMutL-131C but exclusively with MutL-135C that formedcross-links only with a reagent having a long linker arm (Fig. 2,D and E). Of note, both cysteine residues (Cys-162 and Cys-246) are close to peptides in MutS that had been shown byhydrogen/deuterium exchange mass spectrometry to becomeprotected in the presence of MutL (32) (see below).N-terminal Half of MutL Is Sufficient for Cross-link Forma-

tion with MutS—Because bacterial MutS exist in dimer-te-tramer equilibrium, the interpretation of the cross-linking datashown in Table 1 is ambiguous, i.e. which subunit of a MutStetramer is cross-linked to MutL. Previously, we described thegeneration of a fully functional cysteine-free MutS variant thatis unable to form tetramers. In this variant, Asp-835 in thedimerization/tetramerization domain was replaced by an argi-nine residue (8). We generated the single cysteine variantMutS-246C/D835R and tested this dimeric MutS for cross-linking with MutL-131C. Indeed, cross-linking yields werecomparable with that obtained for MutS-246C (i.e. about 50%;Table 1), and cross-linking was dependent on the presence ofDNA (supplemental Fig. S1B). This suggests thatMutL is cross-

linked to only one subunit of the MutS dimer and not to bothsubunits of one dimer in a MutS tetramer. To address whetherthe N-terminal half comprising the ATPase domain of MutLis sufficient for ternary complex formation, we generated a40-kDa single cysteine N-terminal fragment of MutL (residues1–331; LN40-131C). In the absence of nucleotide or in the pres-ence of ADP, this fragment has been shown to exist predomi-nantly in the monomeric form, whereas ATP or AMPPNPbinding supports the formation of dimers (13, 52) (Fig. 3A).When tested for chemical cross-linking toMutS-246C/D835R,we observed high yield cross-linking between MutS and LN40only in the presence of ATP andAMPPNP butmuch less cross-linking (�10%) with ADP (Fig. 3B, lanes 2–4). In the presenceof AMPPNP, an additional band corresponding to the LN40dimer was observed, consistent with the reported stable dimerformation of LN40under these conditions (Fig. 3B, lane 3). Thisband was absent or less pronounced in the presence of ATP orADP (Fig. 3B, lanes 2, 4, and 5). LN40 can form dimers in thepresence ofADPorATP that are not stable (13). In suchdimers,Cys-131 can either cross-link within the LN40 dimer (Fig. 3B,lanes 4 and 5) or, in the presence of MutS-246C, form a cross-link to MutS (Fig. 3B, lane 2). Under this competitive situation(Fig. 3B, lane 2 versus lane 5), the cross-link was formed pref-erentially to MutS, whereas in the absence of MutS, the LN40-LN40 cross-link was observed. In the presence of AMPPNP,stable dimers of LN40 are formed (13). As expected, the effi-ciency for LN40-LN40 cross-link formation was increased (Fig.3B, lane 3).MutS andMutL Variants Are Proficient inMMR in Vivo and

in Vitro—The MMR proficiency of the MutS and MutLmutants used in the study was tested in vivo (Table 1 and Ref.37). All MutS and MutL variants tested in this study showedMMR activity in vivo (�95 or� 90%, respectively). In addition,the variants MutS-WT, MutS-246C/D835R, and MutL-131C,which showed the highest cross-linking yields, were tested in anin vitro mismatch-provoked MutH activation assay using cir-cular DNA containing a GT mismatch and a hemimethylatedGATC site (see “Experimental Procedures” for details). Weobserved similar apparent nicking rates (kobs) of 4.9 and 3.4min�1 for MutL-WT in combination with MutS-WT andMutS-246C/D835R, respectively (Table 2). The activities withMutL-131C both withMutS-WT andMutS-246C/D835Rwerereduced (1.8 and 0.7min�1) but still significantly higher than inthe absence of MutL (�0.1 min�1). These data suggest that theinteraction between these single cysteineMutS andMutL vari-ants is in principle still functional.Modification of MutS and MutL with Fluorescent Dyes—Be-

cause Cys-246 of MutS could be cross-linked to MutL-131C inthe ternary complex, this position must be accessible to allowmodification with the cross-linking reagent. To obtain addi-tional information on the MutL binding of MutS, we modifiedMutS-246C/D835R and for comparison MutS-449C/D835Rusing the fluorescent dyes Alexa Fluor 488- and Alexa Fluor594-maleimide. Both proteins could be modified to �95% asquantified using absorption spectroscopy after removal ofexcess dye by size exclusion chromatography (data not shown).SDS-PAGEanalysis followed by fluorescence imaging andCoo-massie Blue staining confirmed the covalentmodification of the

TABLE 1Summary of cross-linking analyses for MutS variants tested

MutS variant Domaina

Percent cross-linkto MutL-131Cb inpresence of ATP

In vivo activity,cmedian (range)

None N.D. 156 (100–490)WT 40 12 (182) 1 (0–15)CF No 2 (0–5)8C MBD 28 8 (12) 378C MBD 26 7 (9) 1 (0–3)93C MBD 46 10 (20) 3 (0–6)93/239C MBD and connector 48 8 (6) 4 (2–9)103C MBD 30 13 (8) 1162C Connector 16 3 (5) 1 (0–17)239C Connector 9 2 (2) 4 (1–8)246C Connector 62 5 (12) 3 (3–13)246C/D835R Connector 55 8 (28) 5 (3–18)284C Core �5 0 (0–6)297C Core 11 9 (3) 2 (1–4)427C Lever 38 2 (3) 2 (0–8)449C Clamp 21 1 (2) 1 (0–20)483C Clamp 14 1 (3) 1 (0–1)526C Lever 24 1 (2) 3 (0–3)569C ATPase �5 (4) 1 (0–3)686C ATPase �5 (2) 1 (0–9)711C ATPase �5 (2) 5 (1–18)

a Domain containing the cysteine residues: mismatch-binding domain (MBD; resi-dues 1–115), connector domain (residues 116–266), core domain and levers(residues 267–443 and 504–567), clamp domain (residues 444–503), ATPasedomain (residues 568–800) (10).

b Cross-linking with BMPEO�4 was performed with 400 nM MutS variant and1000 nM MutL-131C in the presence of a GT mismatch DNA (100 nM). Inparentheses are the number of independent experiments. N.D., not determined.

c In vivo activity was scored from E. coli TX2929 transformed with the expressionplasmid coding for the indicated MutS variant (38). The numbers of rifampicin-resistant clones plated from 1 ml of overnight culture are given (see Ref. 8 fordetails). Median and ranges are from at least five independent experiments.




http://ww

w.jbc.org/

Dow

nloaded from


http://www.jbc.org/

proteinwith the dye (data not shown). Similarly, we couldmod-ify the single cysteine MutL variant MutL-297C, a functionalsingle cysteine MutL variant that had been studied before (37),with Alexa Fluor 488 to �95%.Sedimentation Velocity Analysis of Fluorophore-modified

MutS andMutL—Todeterminewhether the fluorophoremod-ification had an influence on the quaternary structure of theproteins, we analyzed the proteins modified with Alexa Fluor488 using sedimentation velocity analysis monitoring the fluo-

FIGURE 2. Cys-246 in connector domain forms cross-link to Cys-131 ofMutL-131C. A and B, time course of chemical cross-linking in the presence ofATP using single cysteine MutS variants with a cysteine residue in the mis-match-binding domain (MutS-93C) (A) or in the connector domain (MutS-246C) (B). The samples were analyzed by 6% SDS-PAGE and stained with

colloidal Coomassie Blue. The cross-linking reaction and the concentrationsused are the same as described in Fig. 1. Note that the electrophoretic mobil-ity of the MutS-MutL complex (S x L) is decreased for the MutS-246C-MutL-131C as the branching point in the cross-linked complex moved closertoward the center of the MutS protein sequence. C, quantitative analysis ofthe cross-link reaction kinetics (see “Experimental Procedures” for details).D, cross-linking of MutS-162C to MutL-135C is dependent on the range of thecross-linker. Cross-linking experiments were carried out essentially asdescribed under “Experimental Procedures” with the exception thatBM[PEO]4 was replaced with methanethiosulfonate cross-linkers (MxM) ofvarying spacer arm length, and the reaction was stopped with N-ethylma-leimide. Note that cross-link yield between MutS-162C and MutL-135C (S x L)was only high in the presence of DNA and with the cross-linker M17M with thelongest spacer arm (lane 3). No band corresponding to the MutS-MutL cross-link (S x L) was observed in the absence of either MutS-162C (lane 1) or MutL-135C (lane 2). The cross-link yield of the MutS-MutL cross-link was muchreduced when the length of the cross-linker was shortened (lanes 4 and 5) orDNA was omitted from the reaction (lane 6). Less cross-link formation wasobserved when using BM[PEO]4 or when variants MutS-686C or MutS-449Cwere used instead of MutS-162C (data not shown). E, chemical structures ofthe methanethiosulfonate cross-linkers used.

FIGURE 3. LN40 is sufficient for complex formation and cross-linking withMutS. A, ATPase/dimerization cycle of the N-terminal fragment of MutL(LN40). Upon ATP binding (black sphere), LN40 undergoes a conformationalchange (indicated by dark gray) that allows dimerization. After ATP hydrolysis(light gray), the subunits dissociate and release ADP (gray sphere) (13, 52).B, the N-terminal fragment of MutL containing a single cysteine at position131 (LN40-131C) was tested for complex formation with dimeric MutS-246C/D835R using chemical cross-linking in the presence of the indicated nucleo-tides essentially as described in Fig. 1. The samples were analyzed by 6%SDS-PAGE and stained with colloidal Coomassie Blue. Note that the mono-meric form of LN40 has run out of the gel. S, MutS.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

rescence of the Alexa Fluor 488 dye in the analytical ultracen-trifuge. In the presence of ADP, the c(s) distributions (40)yielded about 90% of the total protein sedimenting with a sed-imentation coefficient s20,w of 8.2 S for MutS-246C/D835R-Alexa Fluor 488 and 8.3 S for MutS-449C/D835R-Alexa Fluor488, respectively, indicative of the formation of MutS dimers(Table 3 and supplemental Fig. S3). In the presence of ATP,s20,w increased slightly to 8.4 S, indicating a slightly more com-pact form of the ATP-bound proteins as has been observed byothermethods (53). Addition of a 42-bpGTheteroduplexDNA(GT42) to the protein in the presence of ADP resulted in anincrease of the uncorrected sedimentation coefficient from 7.9or 8.0 to 8.9 S, showing the formation of a complex betweenMutS and DNA. These data demonstrate that chemical modi-fication of position 246 or 449 in MutS did not affect the qua-ternary structure of the protein and was not inducing aggrega-tion. Moreover, the protein retained its ability to interact withDNA. Similarly, we performed a sedimentation velocity analy-sis withMutL-297Cmodified with Alexa Fluor 488 in the pres-ence of different nucleotides. For the modified MutL-297C inthe presence of AMPPNP, the c(s) analysis yielded about 90% ofthe protein sedimenting in a single boundary with a sedimen-tation coefficient s20,wof 6.1 S, indicating the formation ofMutLdimers under these conditions with a frictional ratio f/f0 �1.53. (Table 2 and supplemental Fig. S3). In the presence ofATP, the protein sedimentedmore slowly (s20,w of 5.8. S and f/f0of 1.62), suggesting that the complex is less compact as com-pared with the one formed with AMPPNP. In the presence ofADP, the c(s) distributions were much broader compared withthose of the other nucleotides and showed several peaks. SDS-PAGE analysis of the samples after centrifugation showed sub-stantial degradation of the modified MutL-297C protein in thepresence of ADP only, indicating that the protein is less stablein the presence of ADP (data not shown). Taken together, thesedata suggest that the modified MutL-297C is not aggregatingand still able to form stable dimers in the presence of AMPPNP

and ATP as has been observed before by size exclusion chro-matography (14) or for the eukaryotic homologues by scanningforce microscopy (54).Ternary Complex Formation of DNA,MutS, andMutLMon-

itored by FRET—To obtain additional information on thestructure of theMutS-MutL complexes, we tested whether ter-nary complex formation can be monitored by FRET usingMutS-246C/D835R modified with Alexa Fluor 594 and MutL-297CmodifiedwithAlexa Fluor 488. Alexa Fluor 488 andAlexaFluor 594 form a FRET pair with a theoretical R0 of 57 Å (55).Incubation of 200 nM MutS with 200 nM MutL in the presenceof ATP did not result in any significant FRET (SR/SG ratio of0.08) (Fig. 4A). This is in agreement with previous observationsthat MutS does not interact with MutL in the absence of DNA.Upon addition of 50 nM 484-bp DNA containing a single GTmismatch, we observed a strong decrease of the donor fluores-

TABLE 2Mismatched-provoked activation of MutH by MutS and MutL variantsMutH incision of a 3.3-kbp DNA containing a GT mismatch and one hemimethyl-atedGATC site wasmonitored using 15 nMDNA, 200 nMMutH, 400 nMMutL, and400 nM MutS at 37 °C. kobs 1 S.E. from global fits are shown (n � 2–4). Back-ground incision rates (in the absence of a mismatch, MutS, or MutL) were �0.1min�1.

Apparent first orderrate constants, kobs

MutL-WT MutL-131C

min�1

MutS-WT 4.9 0.9 1.8 0.3MutS-246C/D835R 3.4 0.6 0.7 0.1

TABLE 3Sedimentation velocity analyses of MutS and MutL variants labeledwith Alexa Fluor 488

Variant

s20,w (frictional ratio f/f0 of the dimer) sexp

ADP ATP AMPPNP �ADP�ADP�GT42

MutS-246C/D835R 8.2 S (1.43) 8.4 S (1.40) 7.9 S 8.9 SMutS-449C/D835R 8.3 S (1.42) 8.4 S (1.40) 8.4 Sa (1.40) 8.0 S 8.9 SMutL-297C N.D.b 5.8 S (1.62) 6.1 S (1.53)a Measurement was done in 25 mM Tris (instead of HEPES) (pH 7.5), 125 mM KCl,5 mM MgCl2, 0.01 mM EDTA, and 1 �M BSA.

b Not determined.

FIGURE 4. Ternary complex formation between MutS and MutL moni-tored using FRET. A and B, fluorescence emission spectra (excitation at 470nm; normalized to the emission at 515 nm) were recorded with MutL-297C(200 nM) labeled with Alexa Fluor 488 in the presence of 1 mM ATP in HT buffer(25 mM HEPES/KOH (pH 7.5), 0.05% Tween 20, 125 mM KCl, and 5 mM MgCl2)(green line). Next 200 nM MutS-246C (A) or MutS-449C (B), respectively, labeledwith Alexa Fluor 594 were added (red) followed by the addition of 50 nM

484-bp GT heteroduplex DNA (orange) and finally KCl to a final concentrationof 300 mM (black dashed line). The inset shows the ratio (SR/SG) between redsignal (SR; emission at 615 nm) and green signal (SG; emission at 515 nm).Averages and standard deviations (error bars) from two independent experi-ments are plotted.




http://ww

w.jbc.org/

Dow

nloaded from



http://www.jbc.org/

cence with a concomitant increase in acceptor fluorescenceindicative of FRET (SR/SG ratio of 0.36). Increasing the concen-tration of KCl to 300 mM, which results in dissociation of MutSfrom DNA, led to a rise in donor and drop in acceptor fluores-cence back to the values obtained in the absence of DNA (SR/SGratio of 0.07). Similar DNA-induced FRET between MutS-449C/D835R-Alexa Fluor 594 andMutL-297C-Alexa Fluor 488was observed; albeit the extent of FRET was reduced (Fig. 4B).These results show that it is possible to monitor the ternarycomplex formation between DNA, MutS, and MutL usingFRET.Moreover, the differences in the FRET efficiency suggestthat position 297 in MutL is closer to position 246 than toposition 449 in MutS and that the distance has to be in therange of R0.MutS Forms Internal Cross-link between Cys-93 in Mis-

match-binding Domain andCys-239 in Connector Domain—Inthe process of identifying the cysteine residue inwild typeMutSthat cross-links toMutL-131C (see above), we noticed that withsomeMutS variants an additional bandwith higher electropho-retic mobility than uncross-linked MutS was observed aftercross-linking (Fig. 5A, lanes 1–4, and B, lanes 3 and 6, andsupplemental Fig. S4). The formation of this cross-link wasdependent on the presence of ATP and DNA (Fig. 5B, lanes3–6) andwas observed both in the absence or presence ofMutL(Fig. 5A, lanes 1–4). Inspection of available crystal structures ofMutS revealed that only several pairs of cysteine residues are ina distance compatible with the length of the cross-linker used(297 and 569, 19 Å; and 235 and 239, 7 Å). However, this cross-link band was observed also with two triple cysteine variantsthat were without these pairs, i.e. MutS-(93/239/297)C andMutS-(93/239/711)C. Distances between cysteine 93 of themismatch-binding domain and cysteine 239 of the connectordomain are 24 and 28 Å (within subunit A or B), respectively,whereas the other pairs show larger distances (38–49 Å) (Fig.5C). To answer whether the observed cross-link can be formedbetween Cys-93 and Cys-239, we generated the double cysteinevariant MutS-(93/239)C, which, similar to the triple cysteinevariant MutS-(93/239/297)C, formed not only a cross-link toMutL-131C but also the internal cross-link (Fig. 5A, lane 1). Nointernal cross-link formation was observed in any of the singlecysteine variants, e.g. MutS-93C (Figs. 1C and 5A, lane 5) orMutS-239C (data not shown). Of note, the formation of theinternal cross-link was ATP-dependent as no internal cross-link formation was observed when ATP was replaced by ADP(Fig. 5B, lane 4), suggesting that an ATP-induced conforma-tional change is required to allow cross-link formation. Thisagrees with the fact that the distance between Cys-93 and Cys-239 (�27.3 Å) observed in the co-crystal structure of MutS incomplex with mismatched DNA in the presence of ADP islarger than the range (�17 Å) of the BM[PEO]4 cross-linker(Fig. 1A).

FIGURE 5. ATP and DNA induce conformational change in MutS bringingmismatch-binding domain closer to connector domain. A, chemical cross-linking of variants MutS-93C, MutS-(93/239)C, and MutS-(93C/239C/297)C inthe presence (lanes 1, 3, and 5) and absence (lanes 2 and 4) of MutL-131C and484-bp GT heteroduplex DNA (GT484). Cross-linking was carried out in thepresence of 1 mM ATP essentially as described in Fig. 1 (see “ExperimentalProcedures” for details). A band (marked by an orange arrowhead) migratingslightly faster than MutS was observed only with MutS-(93/239)C and MutS-(93/239/297)C. B, internal cross-link formation in MutS is ATP- and DNA-de-pendent. Cross-linking between MutL-131C (1 �M) and the triple cysteinevariant MutS-(93/239/297)C (0.4 �M) (lanes 3– 6), MutS-WT (lane 9), and with-out MutS (lane 7) in the presence of 100 nM heteroduplex DNA (GT484), 1 mM

ATP, and 125 mM KCl with BM[PEO]4 (see “Experimental Procedures” fordetails). The reactions in lanes 2 and 8 were done in the absence of BM[PEO]4with the triple cysteine MutS variant and MutS-WT. Reactions were analyzedby 6% SDS-PAGE followed by colloidal Coomassie Blue staining. The blackboxes in lanes 3 and 6 mark the MutS double band. Lane 1, molecular massmarkers. C, positions of cysteine residues in MutS. The subunits of MutS arecolored in gray (subunit A) or according to the domains (subunit B) as in Fig.1B. Cysteine residues are shown as yellow spheres. In addition, positions ofcysteine residues of two single cysteine variants in the connector domain(MutS-162C and MutS-246C) are shown in magenta. The peptides 205–213and 674 – 688 are shown in dark cyan and dark green, respectively. The resi-dues Gln-211/Gln-212 (shown in dark blue) have been shown to be important

for the interaction with MutL (32). Note that the distance between the sulfuratoms of Cys-93 and Cys-239 is 27.3 Å, which is 10 Å larger than the maximumspan of the BM[PEO]4 cross-linker (17.8 Å). The arrow indicates the direction ofmovement of the mismatch-binding domain toward the connector domainupon ATP binding required to allow internal cross-link formation. L, MutL; S,MutS.




http://ww

w.jbc.org/

Dow

nloaded from


http://www.jbc.org/

Models for MutS-MutL Complex—The cross-linking datapresented here and available information from literature (32,37) were used to generated a coarse grained structural modelusing the mismatch-binding and connector domains of MutSand the N-terminal domain of MutL (see supplemental data).The top cluster containing nine models for the complex areshown in Fig. 6A. After reconstruction of MutS, the previouslypublished docking model of theMutL-MutH complex (37) wassuperimposed, and DNA was added from the Haemophilusinfluenzae MutH-DNA crystal structure (56) (Fig. 6B). TheC-terminal domain of MutL had been omitted from modelingbecause the flexible linker between the N-terminal domain ofMutL and the C-terminal domain of MutL makes it difficult todetermine the position of the C-terminal domain of MutL inthis model. In these models, the distance between position 246ofMutS andposition 297 ofMutL (chainAor chainB) is�40Å,whereas the distance between position 449 of MutS and posi-tion 297 of MutL is �50 Å. This fits well to the higher FRETefficiency observed with Alexa Fluor 594-labeled MutS-246C/D835R with Alexa Fluor 488-labeled MutL-297C comparedwith the experiment with Alexa Fluor 594-labeledMutS-449C/D835R with Alexa Fluor 488-labeled MutL-297C (Fig. 4). Ofnote, although not used for filtering during docking, residues205–213, especially Gln-211 and Gln-212, of MutS are in closecontact toMutL (Fig. 6B). In summary, themodel for theMutS-MutL complex is in agreement with the experimental data pre-sented in this work and data published before (32, 37).

DISCUSSION

The complex between MutS and MutL formed on mis-matched DNA in the presence of ATP is an important coordi-nator of DNAmismatch repair. Despite numerous efforts, littleis known about its quaternary structure/arrangement. Thedynamic nature of this complex and the requirement of longDNA for its formation have significantly hampered structuralanalyses in the past. Using site-directed chemical cross-linkingand FRET, we were able to identify regions in MutS and MutLthat are in proximity in the complex. Moreover, cross-linkingenabled us to trap the dynamic complex, which should allowmore detailed structural analyses in the future.In a previous study, we demonstrated that the N-terminal

ATPase domain of MutL is in proximity to MutS (43). A MutLvariant with a single cysteine at position 131 formed a photo-and a chemical cross-link to MutS. Here we showed that thethiol-specific bismaleimide reagent BM[PEO]4 cross-linkedCys-93 in themismatch-binding domain ofMutS to Cys-131 inthe ATPase domain of MutL (Fig. 1). To our knowledge, this isthe first time that a link between themismatch-binding domainand MutL has been shown. The lack of efficient cross-link for-mation in the absence of mismatched DNA or ATP (Fig. 1)suggests that this cross-link reaction is indicative for the ter-nary complex formation. These conclusions are corroboratedby our cross-linking experiments with a MutS variant contain-ing a cysteine residue in the connector domain (MutS-162C

FIGURE 6. Model of complex between MutS, MutL, and MutH on DNA. A, models for the MutS-MutL complex were generated using the Hex program, anddocking decoys were scored using the constraints obtained from the cross-linking data presented in this work (see “Experimental Procedures” and supple-mental information for details). The top nine docking solutions of residues 9 –266 of MutS (mismatch-binding (blue) and connector domains (cyan)) with theN-terminal domain of MutL (MutL-NTD) shown in orange are superimposed. The position of residues 93 (blue), 162, and 246 (both magenta) of MutS andpositions of residues 131 (orange) and 135 (light orange) of MutL are indicated as spheres. B, a structural model for the MutS, MutL, and MutH complex wasgenerated based on the crystal structures of E. coli MutS with a GT mismatch (green; Protein Data Bank code 1e3m) superimposed to one of the docking modelsshown in A and the published complex between the N-terminal domain of E. coli MutL in complex with AMPPNP (orange; Protein Data Bank code 1b63) andE. coli MutH (red; Protein Data Bank code 2azo, chain B) (37). The DNA bound to MutH was taken from the structure of H. influenzae MutH in complex with DNAcontaining a hemimethylated GATC site (gray; Protein Data Bank code 2aor) (56). Coloring for MutS is identical to that in Fig. 5. Residues making chemicalcross-links are indicated by their number and connected by black dashed lines. In this model, the distance between position 297 (chain A or chain B) of MutL toposition 246 (chain B) in MutS is 35 Å. In contrast, the distances to position 449 (chain A or chain B) in MutS are 68 and 52 Å, respectively. Note that the DNA andthe mismatch-binding domain(s) are likely to change their conformation upon ATP binding and complex formation with MutL (indicated by the gray arrow). Agray dashed line indicates how the DNA could run from MutS via MutL to MutH. C, model for the formation of a complex between MutS-MutL and MutHscanning the DNA for the strand discrimination signal. After mismatch binding by MutS, ATP triggers the formation of a long lived sliding clamp. MutL can bindto this clamp. Binding of MutH to MutL leads to the formation of a quaternary complex in which MutS and MutH scan the DNA for a strand discrimination signal,i.e. a hemimethylated GATC site.




http://ww

w.jbc.org/

Dow

nloaded from




http://www.jbc.org/

and MutS-246C) (Table 1 and Fig. 2). Interestingly, cross-link-ing ofMutL-131C to Cys-246 in the connector domain ofMutSwasmuch faster compared with the reaction with Cys-93 in themismatch-binding domain of MutS (Fig. 2C). This could beexplained by the higher solvent accessibility of Cys-246 com-pared with the partially buried Cys-93. However, it might alsoreflect that the mismatch-binding domain becomes mobileafter the ATP-induced conformational change as was proposedbefore (24, 25). In contrast, Cys-246 remains in the same spatialposition to the primary interaction site in the connectordomain, i.e. Gln-211/Gln-212 (Fig. 5B and Ref. 32). A mobilemismatch-binding domain also explains why several variantswith single cysteine residues in this domain were able to forman ATP- and mismatch-dependent cross-link to MutL-131C(Table 1). Moreover, ATP-induced mobilization of the mis-match-binding domain is consistent with the internal cross-link formation between Cys-93 of the mismatch-bindingdomain and Cys-239 in the connector domain that otherwiseare too far separated for cross-linking whenMutS is bound to amismatch in the absence of ATP (Fig. 5).Further experiments showed that the N-terminal fragment

of MutL (LN40-131C) was sufficient for cross-linking withMutS, although it cannot bind to DNA alone (14). This corrob-orates earlier data showing that DNA binding of MutL is notrequired for complex formation withMutS (57). Although full-length MutL has been shown to exist in the presence of ADPpredominantly in an open conformation, it can also adopt aclosed conformation (up to 30%) (14, 52, 54).In the presence of ATP or AMPPNP, the closed conforma-

tion is the dominant form of MutL, and cross-linking to MutSwas enhanced in the presence of these nucleotides (Figs. 1 and2). In contrast, LN40 was not able to form stable dimers in thepresence of ADP andATP (reduced LN40-LN40 cross-link for-mation; Fig. 3B, lanes 2 and 4) due to themissing LC20 domain,which links the two N-terminal domains in the full-lengthMutL (13). However, in the presence of ATP, high yield cross-link formation between MutS-246C/D835R and LN40-131Cwas observed, suggesting that the monomeric form of MutL issufficient for the interaction with MutS (Fig. 3B, lane 2). In thepresence of the non-hydrolyzable ATP analogue AMPPNP,LN40 forms dimers (13), which could interact with and cross-link toMutS (Fig. 3B, lane 3). Our results are in agreement withthe observation that severalmutants ofMutS defective in eitherATP binding, ATP hydrolysis, or dimerization are still capableof forming a ternary complex with MutL on mismatched DNA(58).Taken together, our results suggest that the ATPase domain

ofMutL is in proximity to both themismatch-binding and con-nector domains of MutS. These conclusions are corroboratedby the FRET analysis usingAlexa Fluor 594-labeledMutS-246Cand with Alexa Fluor 488-labeled MutL-297C, indicating thatthese two positions are at a distance of 50–60 Å (Fig. 4A) andthat the observed proximity between these two domains is notan artifact of the cross-linking reaction. In contrast, the lowerFRET between Alexa Fluor 488-labeled MutL-297C and AlexaFluor 594-labeled MutS-449C suggests a longer distancebetween the ATPase domain and the clamp domain. In sum-mary, our data are in agreement with the recent study on the

ternary complex formation of E. coli MutS/MutL and yeastMutS�/MutL� (32). Using time-resolved hydrogen/deuteriumexchange monitored by mass spectrometry, the authors coulddemonstrate that two regions in E. coli MutS displayeddecreased exchange rates in the presence of MutL, i.e. residues205–213 of the connector domain and residues 674–688 of theATPase domain, which are close to residues 246 and 162,respectively (Fig. 5C and Ref. 32). Moreover, the authors coulddemonstrate that the variant MutS-211,2 (Q211S,Q212S) wasstrongly impaired in MutL binding (Fig. 5C).Notably, cross-linking yields in our experiments were often

close to 50% even under conditions using a molar excess ofMutL over MutS. This observation suggests that only one mis-match-binding domain of the homodimericMutS is close to theN-terminal domain of MutL (Fig. 2 and Table 1). The cross-linking yields betweenCys-93 of themismatch-binding domainand Cys-239 of the connector domain also were close to 50%,suggesting, but not proving, that only one subunit is able toundergo the ATP-induced conformational change required forformation (Fig. 5B). These results fit to the observation that inyeast MutS� MSH2 but notMSH6 interacts withMutL�, lead-ing to the conclusion that one subunit of MutS directly recog-nizes themismatch, and the second subunit recruitsMutL (32).It is tempting to speculate that for E. coli MutS subunit B is incontact with MutL, whereas subunit A recognizes the mis-match. Based on comparative normal mode analysis, a move-ment of the mismatch-binding domain, primarily of subunit A(or MSH6 in eukaryotic MutS�), similar to that shown in Fig.5B, has been proposed (26). However, without additionalexperiments, we cannot distinguish whether one or both sub-units are able to form the internal cross-link that indicates amismatch and ATP-inducedmovement of themismatch-bind-ing domain.We were able to generate a coarse grained structural model

of the MutS-MutL complex that is based on experimental data(Fig. 6). This model is in agreement with the proposed interac-tion surface of MutL for MutS (52) and with the identificationof the connector domain of MutS being important and suffi-cient for the interaction with MutL (32). We are aware thatmore experimental data are needed to refine this model and todetermine additional states of the MutS-MutL complex duringtheMMRprocesses. For example, in ourmodel, we did not takeinto account the conformational changes in MutS that hadbeen observed experimentally (Fig. 5 and Refs. 24, 30, and 59)and modeled using normal mode analysis and moleculardynamics simulations (25, 26). These changes, which are likelyto influence the mode of MutS-DNA interaction, may involvethe transition from a bent to an unbent DNA conformation asproposed before (60) and indicated in Figs. 5C and 6B. Despitethis limitation, it is possible to suggest a coarse grained modelfor the MutS, MutL, and MutH complex on DNA (Fig. 6B)using the previous published model for the MutL-MutH com-plex (37). According to themodel presented, it seems likely thata complex comprisingMutS,MutL, andMutH can assemble onthe DNA without the formation of an intervening DNA loop,thereby enabling the complex to scan the DNA for the stranddiscrimination signal (Fig. 6C). This complex would be affected




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

by protein roadblocks on the DNA as has been observed exper-imentally (61).

CONCLUSIONS

The biochemical analyses of the interaction of several MutSvariants with MutL carried out in this study and the data fromliterature support the following conclusions. First, in the ter-nary complex, the mismatch-binding and connector domainsof MutS are in proximity to the ATPase domain of MutL. Sec-ond, the interaction betweenMutS andMutL does not dependon the formation of the closed form of MutL. Third, ATP-in-duced sliding clamp formation is accompanied by an ATP-in-duced movement of the mismatch-binding domain toward theconnector domain. This movement is independent of the bind-ing ofMutL toMutS. Finally, the structuralmodel presented fortheMutS-MutL-MutH complex fits well tomodels for the cou-pling of the mismatch recognition and strand discriminationprocess involving the formation of an active MutS clamp thatrecruits MutL andMutH scanning the DNA for the strand dis-crimination signal (i.e. in E. coli a hemimethylated GATC site(58, 61)).

Acknowledgments—We thank Lidia Litz for excellent technical assis-tance and Flora Groothuizen for critically reading the manuscript.

REFERENCES1. Modrich, P., and Lahue, R. (1996) Annu. Rev. Biochem. 65, 101–1332. Lahue, R. S., Su, S. S., and Modrich, P. (1987) Proc. Natl. Acad. Sci. U.S.A.

84, 1482–14863. Zhang, Y., Yuan, F., Presnell, S. R., Tian, K., Gao, Y., Tomkinson, A. E., Gu,

L., and Li, G. M. (2005) Cell 122, 693–7054. Constantin, N., Dzantiev, L., Kadyrov, F. A., andModrich, P. (2005) J. Biol.

Chem. 280, 39752–397615. Larrea, A. A., Lujan, S. A., and Kunkel, T. A. (2010) Cell 141, 7306. Drummond, J. T., Li, G. M., Longley, M. J., andModrich, P. (1995) Science

268, 1909–19127. Li, G. M., and Modrich, P. (1995) Proc. Natl. Acad. Sci. U.S.A. 92,

1950–19548. Manelyte, L., Urbanke, C., Giron-Monzon, L., and Friedhoff, P. (2006)

Nucleic Acids Res. 34, 5270–52799. Mendillo, M. L., Putnam, C. D., and Kolodner, R. D. (2007) J. Biol. Chem.

282, 16345–1635410. Lamers, M. H., Perrakis, A., Enzlin, J. H., Winterwerp, H. H., deWind, N.,

and Sixma, T. K. (2000) Nature 407, 711–71711. Obmolova, G., Ban, C., Hsieh, P., and Yang, W. (2000) Nature 407,

703–71012. Warren, J. J., Pohlhaus, T. J., Changela, A., Iyer, R. R., Modrich, P. L., and

Beese, L. S. (2007)Mol. Cell 26, 579–59213. Ban, C., and Yang, W. (1998) Cell 95, 541–55214. Guarne, A., Ramon-Maiques, S., Wolff, E. M., Ghirlando, R., Hu, X.,

Miller, J. H., and Yang, W. (2004) EMBO J. 23, 4134–414515. Kosinski, J., Steindorf, I., Bujnicki, J. M., Giron-Monzon, L., and Friedhoff,

P. (2005) J. Mol. Biol. 351, 895–90916. Pillon, M. C., Lorenowicz, J. J., Uckelmann, M., Klocko, A. D., Mitchell,

R. R., Chung, Y. S., Modrich, P., Walker, G. C., Simmons, L. A., Friedhoff,P., and Guarne, A. (2010)Mol. Cell 39, 145–151

17. Malkov, V. A., Biswas, I., Camerini-Otero, R. D., and Hsieh, P. (1997)J. Biol. Chem. 272, 23811–23817

18. Lebbink, J. H., Georgijevic, D., Natrajan, G., Fish, A., Winterwerp, H. H.,Sixma, T. K., and de Wind, N. (2006) EMBO J. 25, 409–419

19. Schofield,M. J., Brownewell, F. E., Nayak, S., Du, C., Kool, E. T., andHsieh,P. (2001) J. Biol. Chem. 276, 45505–45508

20. Yamamoto, A., Schofield, M. J., Biswas, I., and Hsieh, P. (2000) NucleicAcids Res. 28, 3564–3569

21. Gradia, S., Acharya, S., and Fishel, R. (2000) J. Biol. Chem. 275,3922–3930

22. Jeong, C., Cho,W. K., Song, K.M., Cook, C., Yoon, T. Y., Ban, C., Fishel, R.,and Lee, J. B. (2011) Nat. Struct. Mol. Biol. 18, 379–385

23. Gradia, S., Subramanian, D.,Wilson, T., Acharya, S., Makhov, A., Griffith,J., and Fishel, R. (1999)Mol. Cell 3, 255–261

24. Lamers, M. H., Georgijevic, D., Lebbink, J. H., Winterwerp, H. H., Agi-anian, B., de Wind, N., and Sixma, T. K. (2004) J. Biol. Chem. 279,43879–43885

25. Mukherjee, S., Law, S. M., and Feig, M. (2009) Biophys. J. 96, 1707–172026. Mukherjee, S., and Feig, M. (2009) Biophys. J. 96, L63–6527. Mendillo, M. L., Mazur, D. J., and Kolodner, R. D. (2005) J. Biol. Chem.

280, 22245–2225728. Lebbink, J. H., Fish, A., Reumer, A., Natrajan, G., Winterwerp, H. H., and

Sixma, T. K. (2010) J. Biol. Chem. 285, 13131–1314129. Polosina, Y. Y., and Cupples, C. G. (2010) BioEssays 32, 51–5930. Mendillo, M. L., Putnam, C. D., Mo, A. O., Jamison, J. W., Li, S., Woods,

V. L., Jr., and Kolodner, R. D. (2010) J. Biol. Chem. 285, 13170–1318231. Hargreaves, V. V., Shell, S. S.,Mazur, D. J., Hess,M. T., andKolodner, R. D.

(2010) J. Biol. Chem. 285, 9301–931032. Mendillo, M. L., Hargreaves, V. V., Jamison, J. W., Mo, A. O., Li, S., Put-

nam, C. D., Woods, V. L., Jr., and Kolodner, R. D. (2009) Proc. Natl. Acad.Sci. U.S.A. 106, 22223–22228

33. Thomas, E., Pingoud, A., and Friedhoff, P. (2002) Biol. Chem. 383,1459–1462

34. Heinze, R. J., Giron-Monzon, L., Solovyova, A., Elliot, S. L., Geisler, S.,Cupples, C. G., Connolly, B. A., and Friedhoff, P. (2009)Nucleic Acids Res.37, 4453–4463

35. Kirsch, R. D., and Joly, E. (1998) Nucleic Acids Res. 26, 1848–185036. Toedt, G. H., Krishnan, R., and Friedhoff, P. (2003) Nucleic Acids Res. 31,

819–82537. Giron-Monzon, L., Manelyte, L., Ahrends, R., Kirsch, D., Spengler, B., and

Friedhoff, P. (2004) J. Biol. Chem. 279, 49338–4934538. Feng, G., and Winkler, M. E. (1995) BioTechniques 19, 956–96539. Pace, C. N., Vajdos, F., Fee, L., Grimsley, G., and Gray, T. (1995) Protein

Sci. 4, 2411–242340. Schuck, P. (2000) Biophys. J. 78, 1606–161941. Durchschlag, H. (1986) in Thermodynamic Data for Biochemistry and

Biotechnology (Hinz, H. J., ed) pp. 45–128, Springer-Verlag, Berlin42. Ritchie, D. W., and Kemp, G. J. (2000) Proteins 39, 178–19443. Plotz, G., Welsch, C., Giron-Monzon, L., Friedhoff, P., Albrecht, M., Pi-

iper, A., Biondi, R. M., Lengauer, T., Zeuzem, S., and Raedle, J. (2006)Nucleic Acids Res. 34, 6574–6586

44. Jacobs-Palmer, E., and Hingorani, M. M. (2007) J. Mol. Biol. 366,1087–1098

45. Zhai, J., and Hingorani, M. M. (2010) Proc. Natl. Acad. Sci. U.S.A. 107,680–685

46. Blackwell, L. J., Wang, S., and Modrich, P. (2001) J. Biol. Chem. 276,33233–33240

47. Baitinger, C., Burdett, V., and Modrich, P. (2003) J. Biol. Chem. 278,49505–49511

48. Junop, M. S., Obmolova, G., Rausch, K., Hsieh, P., and Yang, W. (2001)Mol. Cell 7, 1–12

49. Natrajan, G., Lamers, M. H., Enzlin, J. H., Winterwerp, H. H., Perrakis, A.,and Sixma, T. K. (2003) Nucleic Acids Res. 31, 4814–4821

50. Green, N. S., Reisler, E., andHouk, K. N. (2001) Protein Sci. 10, 1293–130451. Iyer, R. R., Pluciennik, A., Burdett, V., andModrich, P. L. (2006)Chem. Rev.

106, 302–32352. Ban, C., Junop, M., and Yang, W. (1999) Cell 97, 85–9753. Kato, R., Kataoka,M., Kamikubo, H., and Kuramitsu, S. (2001) J. Mol. Biol.

309, 227–23854. Sacho, E. J., Kadyrov, F. A., Modrich, P., Kunkel, T. A., and Erie, D. A.

(2008)Mol. Cell 29, 112–12155. Stevens, N., Dyer, J., Martí, A. A., Solomon, M., and Turro, N. J. (2007)

Photochem. Photobiol. Sci. 6, 909–91156. Lee, J. Y., Chang, J., Joseph, N., Ghirlando, R., Rao, D. N., and Yang, W.




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

(2005)Mol. Cell 20, 155–16657. Junop, M. S., Yang, W., Funchain, P., Clendenin, W., and Miller, J. H.

(2003) DNA Repair 2, 387–40558. Acharya, S., Foster, P. L., Brooks, P., and Fishel, R. (2003) Mol. Cell 12,

233–24659. Lamers, M. H., Winterwerp, H. H., and Sixma, T. K. (2003) EMBO J. 22,

746–75660. Wang, H., Yang, Y., Schofield,M. J., Du, C., Fridman, Y., Lee, S. D., Larson,

E. D., Drummond, J. T., Alani, E., Hsieh, P., and Erie, D. A. (2003) Proc.Natl. Acad. Sci. U.S.A. 100, 14822–14827

61. Pluciennik, A., and Modrich, P. (2007) Proc. Natl. Acad. Sci. U.S.A. 104,12709–12713




http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Annet Reumer, Ute Curth, Titia K. Sixma and Peter FriedhoffInes Winkler, Andreas D. Marx, Damien Lariviere, Roger J. Heinze, Michele Cristovao,

Escherichia coliMismatch Repair in Chemical Trapping of the Dynamic MutS-MutL Complex Formed in DNA

doi: 10.1074/jbc.M110.187641 originally published online March 15, 20112011, 286:17326-17337.J. Biol. Chem.

10.1074/jbc.M110.187641Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

http://www.jbc.org/content/suppl/2011/03/29/M110.187641.DC1

http://www.jbc.org/content/286/19/17326.full.html#ref-list-1

This article cites 60 references, 21 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M110.187641

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;286/19/17326&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/286/19/17326

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=286/19/17326&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/286/19/17326

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/suppl/2011/03/29/M110.187641.DC1

http://www.jbc.org/content/286/19/17326.full.html#ref-list-1

http://www.jbc.org/

VOLUME 286 (2011) PAGES 5197–5203DOI 10.1074/jbc.A110.161927

The multicompartmental p32/gClqR as a new target forantibody-based tumor targeting strategies.David Sanchez-Martín, Angel M. Cuesta, Valentina Fogal, Erkki Ruoslahti,and Luis Alvarez-Vallina

PAGE 5203:

Ref. 7 should read as follows: Buchsbaum, D. J., Chaudhuri, T. R.,Yamamoto, M., and Zinn, K. R. (2004) Ann. Nucl. Med. 18, 275–283.

VOLUME 286 (2011) PAGES 17326 –17337DOI 10.1074/jbc.A110.187641

Chemical trapping of the dynamic MutS-MutL complexformed in DNA mismatch repair in Escherichia coli.Ines Winkler, Andreas D. Marx, Damien Lariviere, Laura Manelyte,Luis Giron-Monzon, Roger J. Heinze, Michele Cristovao, Annet Reumer,Ute Curth, Titia K. Sixma, and Peter Friedhoff

PAGE 17326:

Drs. LauraManelyte and Luis Giron-Monzon have been added to theauthor list. The complete author list is shown above. Their affiliation isthe Institute for Biochemistry, FB 08, Justus Liebig University, D-35392Giessen, Germany.PAGE 17336:

Ref. 2 should read as follows: Lahue, R. S., Au, K. G., and Modrich, P.(1989) Science 245, 160–164.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 25, p. 22706, June 24, 2011© 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

22706 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 286 • NUMBER 25 • JUNE 24, 2011

ADDITIONS AND CORRECTIONS This paper is available online at www.jbc.org

We suggest that subscribers photocopy these corrections and insert the photocopies in the original publication at the location of the origi-nal article. Authors are urged to introduce these corrections into any reprints they distribute. Secondary (abstract) services are urged tocarry notice of these corrections as prominently as they carried the original abstracts.

› content › 286 › 19 › 17326.full.pdf · chemicaltrappingofthedynamicmuts-mutlcomplex...

Documents