NMR Characterization of the Interaction of the Endonuclease Domain of MutL with Divalent Metal Ions and ATP Ryota Mizushima1, Ju Yaen Kim1, Isao Suetake1, Hiroaki Tanaka1, Tomoyo Takai1, Narutoshi Kamiya1, Yu Takano1, Yuichi Mishima1, Shoji Tajima1, Yuji Goto1, Kenji Fukui2¤, Young-Ho Lee1* 1InstituteforProteinResearch,OsakaUniversity,Suita,Osaka,Japan,2RIKENSPring-8Center,HarimaInstitute,Sayo-cho,Sayo-gun,Hyogo,Japan Abstract MutLisamulti-domainproteincomprisinganN-terminalATPasedomain(NTD)andC-terminaldimerizationdomain(CTD), connected with flexible linker regions, that plays a key role in DNA mismatch repair. To expand understanding of the regulationmechanismunderlyingMutLendonucleaseactivity,ourNMR-basedstudyinvestigatedinteractionsbetweenthe CTD of MutL, derived from the hyperthermophilic bacterium Aquifex aeolicus (aqMutL-CTD), and putative binding molecules. Chemical shift perturbation analysis with the model structure of aqMutL-CTD and circular dichroism results revealed that tight Zn2+ binding increased thermal stability without changing secondary structures to function at high temperatures.Peakintensityanalysisexploitingtheparamagneticrelaxationenhancementeffectindicatedthebindingsite for Mn2+, which shared binding sites for Zn2+. The coexistence of these two metal ions appears to be important for the function of MutL. Chemical shift perturbation analysis revealed a novel ATP binding site in aqMutL-CTD. A docking simulation incorporating the chemical shift perturbation data provided a putative scheme for the intermolecular interactions between aqMutL-CTD and ATP. We proposed a simple and understandable mechanical model for the regulation of MutL endonuclease activity in MMR based on the relative concentrations of ATP and CTD through ATP binding-regulated interdomain interactionsbetween CTDand NTD. Citation:MizushimaR,KimJY,SuetakeI,TanakaH,TakaiT,etal.(2014)NMRCharacterizationoftheInteractionoftheEndonucleaseDomainofMutLwith DivalentMetalIonsandATP.PLoSONE9(6):e98554.doi:10.1371/journal.pone.0098554 Editor:FreddieSalsbury,Jr,WakeForestUniversity,UnitedStatesofAmerica ReceivedJanuary1,2014;AcceptedMay5,2014;PublishedJune5,2014 Copyright:(cid:2)2014Mizushimaetal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermits unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited. Funding:ThisworkwassupportedbytheJapaneseMinistryofEducation,Culture,Sports,ScienceandTechnology.Thefundershadnoroleinstudydesign,data collectionandanalysis,decisiontopublish,orpreparationofthemanuscript. CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist. *E-mail:[email protected] ¤Currentaddress:TransGenicInc.KobeResearchInstitute,Chuo-ku,Kobe-shi,Hyogo,Japan Introduction endonuclease activities of human and yeast MutLa were demonstrated in the presence of divalent metal ions such as DNA mismatch repair (MMR) is the process by which post- Mn2+[8,9]. replicative base pair errors are rectified, thereby enhancing the Since MutL lacks base specificity as an endonuclease, under- fidelityofDNAreplication100–1000fold[1].MutationsinMMR standingtheregulationmechanismofMutLendonucleaseactivity genes have been shown to cause hereditary non-polyposis is one of the central themes that has been focused on to expand colorectalcancer,referredtoasLynchSyndrome[2,3].Although knowledgeonMMR.MutLconsistsofaC-terminaldimerization significant differences have been reported between Escherichia coli domain(CTD)andN-terminalATPasedomain(NTD)connected (E. coli) and eukaryotic MMR, the basic concepts of MMR have with disordered linker regions [10,11]. MutL has been shown to beendevelopedprimarilythroughresearchonE.coli[4,5].MMR form heterodimers and homodimers in eukaryotes and prokary- startsfromtherecognitionofmismatchedbasepairsbyMutS,and otes,respectively [5]. is followed by the recruitment of MutL by the mismatch:MutS Alterations in the quaternary structure of MutL were assumed complex. to result from interdomain interactions between the NTD and Themismatch:MutS:MutLcomplexhasbeenshowntoinitiate CTD/NTD, and the NTD may be involved in regulating downstream repair machinery in both E. coli and eukaryotes, in endonuclease activity when it binds with metal ions or ATP which newly synthesized DNA strands are specifically nicked at [11].IsolatedaqMutL-NTDwaspreviouslyshowntostimulatethe either side of the mismatch [6,7]. However, the mechanisms endonuclease activity of aqMutL-CTD in a zinc-dependent underlyingdaughterstrandrecognitionandincisionaredecisively manner [12]. A recent study also indicated that MutS coupled different between E. coli and eukaryotes. While E. coli MMR with the endonuclease activity of MutL stimulated by the DNA possesses MutH endonuclease, which cleaves daughter strands at mismatch inan ATP-dependent manner, and alsothat theNTD hemi-methylated GATC sites, eukaryotes and most bacteria lack of MutL primarily mediated the interaction between MutL and MutH; hence, the counterpart of MutH in eukaryotic and MutS.Therefore,MutSmaystimulatetheendonucleaseactivityof eubacteria species was a topic for speculation until the latent MutL through interdomain interactions between the CTD and PLOSONE | www.plosone.org 1 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity NTD [13]. However, the functional details associated with Sequential Backbone Assignment of aqMutL-CTD changes in the quaternary structure remain largely unknown aqMutL-CTD samples uniformly labeled with 2H-13C-15N, [11,12,14,15]. 13C-15N,or15Nwerepreparedusing50mMpotassiumphosphate Therefore, the intermolecular interactions between MutL and buffer (pH 6.8) containing 100mM KCl, 5mM DTT, 1 mM variousbindingmoleculessuchasdivalentmetalions,ATP,other EDTA,10%D O,and0.02%NaN .Potassiumphosphatebuffer 2 3 MMRproteins,andDNAbasedonMutLconformationsneedto was used to assign aqMutL-CTD because the dispersion and be determined in order to examine the regulation of MutL sharpness of the spectra obtained from samples in potassium endonuclease activity. Recent structural studies using X-ray phosphatebufferwerebetterthanthoseinTris-HClbuffer(Figure crystallography have provided information on CTD conforma- S1). All NMR measurements for backbone assignments were tionsinBacillussubtilisMutL(bsMutL).Thestaticcrystalstructures performedat40uCinanAVANCEII-800spectrometerequipped of the CTD:Zn2+ complex from Neisseria gonorrhoeae and the with a cryogenic probe (Bruker BioSpin, Germany). Data were complex of CTD from Saccharomyces cerevisiae, with fragments processedbyNMR-Pipe[19]andanalyzedbySparky.Sequential containing the MIP-box motifs of the Exo1 and Ntg2 proteins backboneassignmentsofaqMutL-CTDwereperformedprimarily [1,16,17], revealed the detailed interacting sites and binding onuniformly15N-13C-labeledaqMutL-CTD usingasetofthree- modes of theCTD forpartner molecules. dimensional(3D)heteronuclearcoherencetransfermeasurements However, the moderate and/or weak intermolecular interac- ofHNCO,HNCACO,CBCACONH,andHNCACB.Histidine- tions and conformational dynamics of MutL, which are key to specificisotopelabelingwasintroducedbecauseseveralunassigned understanding MMR in solution, have been difficult to detect by peaks,especiallyaroundH400andH404inthea2-b4loopregion, X-ray crystallography. Therefore, solution-state NMR is a hampered the complete sequential backbone assignment (Figure powerful approach used to examine the intermolecular interac- S2).Thismethodallowedustoassignallthreehistidineresidues: tionsbetweenMutLandothermoleculesbecauseNMRprovides H353, H400, and H404. The remaining unassigned peaks could residue-based information, even on weak intermolecular interac- be sequentially assigned based on the assignment information of tionsand proteindynamics insolution. histidine residues and TROSY-HNCACB and TROSY-CBCA- We here described an NMR investigation of the CTD of CONH measurements of2H-15N-13C-labeledsamples. aqMutL-CTD,anditsintermolecularinteractionswithmolecules Assignment results were confirmed by the 1H-15N HSQC suchasZn2+,Mn2+,ATP,andDNA,whichhavebeensuggested spectra of inversely- and arginine/lysine specifically-labeled to interact with the CTD. Based on the almost complete NMR samples (Figure S3). The final assignment rate was ,97%, backbone assignment of aqMutL-CTD (,97% of 103 assignable excluding the 7 proline residues and N-terminal residue. residues),themodelstructureofaqMutL-CTD,moleculardocking Unassigned residues were L317, D335, L397, N398, and R406, simulations, and NMR titration/relaxation measurements, we whichwerelocatedintheunstructuredloopregionsofthemodel revealed the intermolecular interaction and binding sites of structureofaqMutL-CTD(Figure1A).Theassignmentresultwas aqMutL-CTD for binding partners. We further addressed the depicted on the 1H-15N HSQC spectrum of aqMutL-CTD biological implications of Zn2+ and Mn2+ binding and a possible (Figure 1A) and final resonance assignments were deposited to mechanical model for the regulation of MutL endonuclease BioMagRes DB(BMRB accessionnumber 11545). activity. NMR Relaxation Measurement of aqMutL-CTD and Materials and Methods Model-free Analysis 15N longitudinal and transverse relaxation times (T and T ) Protein Expression and Purification and 1H-15N steady state NOE values were measured1 using 2a aqMutL-CTD was expressed and purified as previously 0.4 mM aqMutL-CTD sample in the same buffer used in the described [12]. E. coli was cultured in minimal media containing backbone assignment of aqMutL-CTD using a Bruker DRX-500 13C-glucose and 15N-NH Cl as the sole carbon and nitrogen 4 spectrometer equipped with shielded triple-axis gradient triple- sources,respectively.Thecellbodywascollectedbycentrifugation resonance probes at 40uC (Figure S4). T was measured with 1 and lysed by sonication in 50mM Tris-HCl (pH 7.8) containing relaxationdelaysof5,150,300,450,600,800,1000,1200,1400, 500mMNaCl.Aftercentrifugationat18,000rpmfor60min,the 1600, and 1800ms. T was measured with relaxation delays of 2 supernatantwasheatedat70uCfor10min.Theheatedsolution 7.2,21.6,43.2,64.8,86.4,108.0,129.6,151.2,180.0,208.8,and was again centrifuged and the supernatant was loaded to the 237.0 ms. Single exponential curve fitting was performed using DEAE column (TOSOH, Tokyo, Japan). The pass solution was Sparky to obtain the values of T and T . 1H-15N NOE values 1 2 then loaded onto the SP-sepharose column (GE-Healthcare were calculated by comparing the peak intensities with and Biosciences, USA) pre-equilibrated with 50mM Tris-HCl buffer without 1H saturation of 3 s. The global tumbling time (t ) of m (pH 7.8) containing 1 mM dithiothreitol (DTT). Ammonium aqMutL-CTDwasestimatedtobe11.7 nsusingtheT /T values. 1 2 sulfate was added to the collected fraction to yield a final By incorporating the three relaxation parameters (T , T , and 1 2 concentration of 1M. The solution was again loaded onto the NOE)andt valuetothemodel-freeanalysiswithTensor2[20], m Toyopearl-phenylcolumn(TOSOH,Tokyo,Japan).Thefractions the values of the order parameters (S2) were obtained under the of aqMutL-CTD were collected and concentrated using an assumption ofisotropic tumblingof theCTD. Amicon Ultra centrifugal filter (Millipore, Billerica, USA). The samplesolutionwasthenappliedtothesuperdex75HRcolumn NMR Titrations of aqMutL-CTD with Various Molecules (GE Healthcare Biosciences, USA) pre-equilibrated with 50mM Changes in the NMR signals of the 1H-15N HSQC spectra of potassium phosphate buffer (pH 6.8) containing 100mM KCl, uniformly 15N-labled CTD with the addition of other molecules 1 mM DTT, and 1 mM EDTA. The fractions of aqMutL were (Zn2+,Mn2+,ATP,andADP)weremonitoredusingtheAVANCE collectedandtheirpuritieswerecheckedbySDS-PAGE(.95%). II-800 or AVANCE III-950 spectrometer (Bruker BioSpin, Protein concentrations were determined by the molar extinction Germany) equipped with a cryogenic probe at 40uC. Measure- coefficients of aqMutL,19005M21cm21 at 278nm [18]. mentswereperformedusing0.1mMCTDin50mMTris-HClor PLOSONE | www.plosone.org 2 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity Figure1.Thetwo-dimensionalNMRspectrumandmodelstructureofaqMutL-CTD.(A)Theresultsofthesequentialbackboneassignment ofaqMutL-CTD(aqMutL )aredisplayedinthe1H-15Nheteronuclearsinglequantumcoherence(HSQC)spectrum.Cross-peaksarelabeledusing 316–425 theone-letteraminoacidcodeandresiduenumber.Thealiasedpeaks(G332andD351)onthe15Naxisareindicatedwith0{0.(B)Themodelstructure of the homodimer of aqMutL-CTD was generated by Modeller (see the Materials and Methods). The a-helix and b-strand and both termini are labelled.(C)Thephi(sphere)andpsi(diamond)anglesandorderparameters(S2),ontheleftandrightaxes,respectively,areplottedagainstthe residuenumber.Thea-helix(arrow)andb-strand(rectangle)predictedusingTALOS+(open)andModeller(closed)areshownabove.(D)Thevalues ofS2obtainedfromthemodel-freeanalysis(opencircle)andfromTALOS+(solidcircleplusline)areplottedagainsttheresiduenumbers. doi:10.1371/journal.pone.0098554.g001 PLOSONE | www.plosone.org 3 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity potassium phosphate buffer (pH 7.0) containing 100mM KCl, identity of the primary sequence (,32%) between aqMutL-CTD 1 mMDTT,0.02%NaN ,and10% D O. andbsMutL-CTD. 3 2 The titration of Zn2+ was performed using the 25mM ZnCl AgraphicalrepresentationoftheputativeMn2+bindingsiteof 2 stock solution. Aliquots of the stock solution were added to the aqMutL-CTD, based on the NMR peak intensity analysis, was CTDsolutiontoreachtheratiosof[ZnCl ]/[CTD]=0,0.4,0.8, depicted by referencing the previously published X-ray crystal 2 2.0, and 2.4. Similarly, 50mM MnCl of the stock solution was structure of Mn2+-bound protein (PDB ID: 1N8F), which was 2 prepared, and aliquots of the stock solution were added to shown to possess a similar coordination geometry for Mn2+ aqMutL-CTDinphosphatebuffertoreachtheratiosof[MnCl ]/ bindingusingthe3D-motifsearchcapabilityofSwissPdb-Viewer 2 [aqMutL-CTD]=0,0.2,0.4,0.8,2.0,and4.0.Thepeakintensity [23]. ratiowascalculatedusingtheintensityofthepeakswithoutMnCl 2 asareference.Thestocksolutionof100mMATPandADPwas Docking Simulation of the Complex of CTD with ATP prepared, and aliquots of the stock solution were added to the A docking simulation was performed using myPresto/Sievgene CTD solution to reach the ratios of [ATP]/[CTD] or [ADP]/ based on the results of the NMR chemical shift perturbation [CTD]=0,20,40,60,80,and100.TheexcessadditionofATP analysis of ATP binding to aqMutL-CTD [24]. The ligand over the ratio of 120 resulted in the precipitation of the protein molecule was prepared in Sybyl mol2 file format. The atomic sample. Synthesized 21-mer singled-stranded DNA, 59- charges of ATP were obtained from the following web site: GCGGTCATAGTCAAGATACCG-39 (Integrated DNA Tech- [http://www.pharmacy.manchester.ac.uk/bryce/amber/]. The nologies, Inc.) was annealed to complementary single-stranded force field parameters for the protein were similar to those in DNA (Integrated DNA Technologies, Inc.) to obtain 21-bp AMBER parm99 [25]. double-stranded DNA (dsDNA) and the stock solution of 6mM 21-bpdsDNAwaspreparedaspreviouslydescribed[21].Aliquots Circular Dichroism (CD) Spectroscopy of the stock solution of dsDNA were added to 0.1 mM CTD in Far-UVCDspectraof20mM(0.27 mgml21)aqMutL-CTDin 25mM Tris-HCl buffer containing 25mM KCl, 1 mM EDTA, 50mMTris-HCl(pH7.0)including100mMKClwererecorded 1 mM DTT, 0.02% NaN3, and 10% D2O to reach the ratios of in the absence and presence of 40mM Zn2+, 80mM Mn2+, and [dsDNA]/[CTD]=0, 4,8,and 12. 2 mMATPat40uC,respectively.HeatscanningofaqMutL-CTD Chemical shift differences in the cross-peaks by titration were from 40 to 95–100uC was also performed in the absence and calculated using therelationship: presence of divalent metal ions and ATP under the same concentration settings described above by monitoring CD signals Dd ~hðDd Þ2zð0:158(cid:2)Dd Þ2i0:5 at 220nm at a rate of 1uC min21. CD measurements were tot HN N performed with a J-720 spectropolarimeter (Jasco, Japan) using a cell with a light path of 1mm. CD signals between 195 and where DdHN and DdN are changes in the 1H and 15N chemical 250nmwereexpressedasthemeanresidueellipticity[h](degcm2 shiftsinppm,respectively.Theweightingfactorof0.158wasused dmol21). Temperature was regulated using a PTC-423L Peltier- to adjust the relative magnitudes of the amide nitrogen chemical unit (Jasco,Japan). shift range andtheamide proton chemicalshift range. Isothermal Titration Calorimetry (ITC) Measurement Model Structure of aqMutL-CTD and Its Complexes with CTD solution was dialyzed against 50mM Tris-HCl (pH 7.0) Zn2+ and Mn2+ containing 100mM KCl and 1 mM b-mercaptoethanol which The model structure of aqMutL-CTD was built using the usedinsteadofDTTforminimizingbaselinedriftofthermogram. homologymodelingmethodologybyModeller9.12[22].TheX- ATDandADPweredissolvedbyusingthebufferafterdialysisand raycrystalstructuresofbsMutL(PDBID:3KDG)anditscomplex their concentrations were determined by UV-vis absorbance. All with Zn2+ (PDB ID: 3KDK) were used as a template structure, of the samples were degassed for 3 min at 40uC before being with the secondary structure restraints obtained from the loaded into the calorimeter. Calorimetric experiments were TALOS+secondarystructurepredictionbasedonNMRchemical performed with a VP-ITCinstrument (GE-Healthcare Bioscienc- shift assignment values. es, USA) at 40uC. The nucleotide (2 mM ATP or ADP) in the We further generated the model structures of aqMutL using injection syringe was titrated into 16mM CTD in the ITC cell bacterial NgoL MutL-CTD homodimer (PDB ID: 3NCV) and (FigureS8).Titrationexperimentsconsistedof39injectionsspaced eukaryotic PMS1-CTD in the MutLa-CTD heterodimer (PDB- atintervalsof400sec.Theinjectionvolumewas7 mlandthecell ID:4E4W)foratemplateX-raystructuretoassessthereliabilityof was continuously stirredat 242rpm. the homology modeling (Figure S10). Then, we compared these two model structures with the model structure generated using Light Scattering Measurements bsMutL-CTD. The overall conformation of the model structure Light scattering of all of the sample solutions was measured at with the bacterial X-ray structure is quite similar to that with 40uC using a Hitachi fluorescence spectrophotometer F4500 bsMutL-CTD except for a few residues in the a-helix (a2 in the (Hitachi,Tokyo,Japan)withtheexcitationwavelengthat350nm model structure based on bsMutL-CTD) and the b-strand (b3 in (Figure S9). CTD and ATP were dissolved in 50mM Tris-HCl themodelstructurebasedonbsMutL-CTD).Theslightstructural (pH7.0)containing100mMKCland1mMb-mercaptoethanol. differences in the loop structures were also found. On the one TheamyloidfibrilsofAb peptideswerealsopreparedforthe hand, the other model structure generated with the eukaryotic 1–40 controls of light scattering of aggregates. Ab peptides were X-raystructurecriticallydifferswiththelackofthea-helixinthe 1–40 dissolvedandusedfortheformationofamyloidfibrilsasdescribed C-terminusofCTD(a4inthemodelstructurebasedonbsMutL- in the previous study [26]. The formation of Ab fibrils was CTD) and the additional a-helical component in the N-terminus 1–40 confirmedbythefar-UVCDspectroscopyandthioflavinTassay of CTD. The model structure of aqMutL-CTD generated using (datanot shown). bsMutL-CTD was selected for this study due to the highest PLOSONE | www.plosone.org 4 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity Results Characterization of aqMutL-CTD Stereostructures The 1H-15N HSQC spectrum displayed well-dispersed cross peaks, which indicated that a homodimer of aqMutL-CTD is a solid 3D structure at pH 6.8 and 40uC (Figure 1A). The far-UV CD spectrum of aqMutL-CTD showed a minimum from ,210nmto,220nm,indicatingstructuredsecondarystructures (Figure2A).AhomologymodelofanaqMutL-CTDconformation wasproducedusingModeller9.12toobtaindetailedinformation onthetertiarystructure(Figure1B)[22].bsMutL-CTD(PDBID: 3KDG) [1] was used as the template conformation for modeling because bsMutL-CTD showed ,32% sequence identity to aqMutL-CTD, which was sufficient for reliable homology modeling [27]. The amino acid sequence and conserved motifs ofvariousMutLfromdifferentspeciesareshown(FigureS5).The homology model was generated under the constraints of the secondary structures predicted by TALOS+ [28], estimating dihedral angles based on the assigned chemical shift information of C , Ca, C , HN, and 15N of aqMutL-CTD (Figure 1C). The O b producedmodelofaqMutL-CTDformedaninvertedhomodimer with an interface consisting of four b-sheets (b1–b4), each abundantly incorporating hydrophobic residues, and one a-helix (a3), domain swapping with each other in the aqMutL-CTD dimer(Figure1B).Helixa1,thea1–a2loop,andthea2–b4loop included the conserved motifs DQHA(X) E(X) , ACRISV, and 2 4 CPHGRPI, respectively. The S2 value ranged from 0 to 1 depending on the degree of flexibilityandhigherS2valuesreflectedlowerflexibilityinthefast timescale.Thus,S2couldbeareporterforproteindynamicsand structure.HighS2values(over0.7)inthea-helixandb-strandand lowS2values(below0.7)inunstructuredregionsoftheloopsand terminal parts were consistent with the predicted secondary structuralelements(Figure1D).Theloopbetweenthea1anda2 helices showed high S2 values of approximately 0.8, which indicated a rigidbackbone conformation. Zn2+-Bound aqMutL-CTD Displayed Enhanced Thermostability Zn2+ was previously suggested to be involved in the regulation Figure 2. CD measurements of aqMutL-CTD in the presence ofMutLendonucleaseactivityforproperMMRinvitroandinvivo and absence of binding partners. (A) The far-UV CD spectra of aqMutL-CTDwithout(black)andwithZn2+(red),Mn2+(purple),andATP [1,8,9]andisrelatedtotheinterdomaininteractionsbetweenthe (green) are shown. Noisy data points in the shorter wavelength from CTD andNTD [12,15]. 220nminthepresenceofATPwereomitted.(B)HeatscansofaqMutL- The NMR-based titration of Zn2+ was performed to examine CTD are shown without (black) and with Zn2+ (red), Mn2+ (blue), and the intermolecular interactions between Zn2+ and aqMutL-CTD ATP(green)usingtheCDsignalat220nm. insolution.Weobservedlargechangesinthepeaksofthe1H-15N doi:10.1371/journal.pone.0098554.g002 HSQCspectrum:manynewpeaksappearedprimarilybecauseof slowexchangesandsimultaneouslybroadenedduetointermediate structures (Figure 2A). The thermal denaturation of CTD in the exchanges; however, we could not assign all new peaks. presenceandabsenceofZn2+wasexaminedbytracingCDsignals Representative residues in the slow exchange regime, C402 and at 220nm (Figure 2B). The results of thermal denaturation G405, are indicated by rectangles (Figures 3A and 3B). These experiments demonstrated that Zn2+ binding significantly en- results suggested that Zn2+ binding was tight and related to the hancedthethermostabilityofaqMutL-CTD.FreeCTDbeganto relativelybroadregionsofaqMutL-CTD,whichmaybeinvolved melt cooperatively at ,50uC and was almost completely in a conformational change. The putative Zn2+ binding sites of denatured at ,95uC. Meanwhile, Zn2+-bound CTD largely aqMutL-CTDindicatedbythereportedX-raycrystallographyof changedthethermaldenaturationprofile.CTDmeltedgradually bsMutL-CTD were also illustrated (Figure 3C). The modeled from ,65uC, and heat denaturation was not finished even at complex structure of the CTD and Zn2+ provided putative ,95uC. residuesthatinteractedwiththetwoZn2+ions:H353andE357in DQHA(X) E(X) of the a1 helix, C371 in ACRISV, and C402, 2 4 Determination of the Mn2+ Binding Sites of aqMutL-CTD H404, andR406intheCPHGRPI motif. in Solution Thefar-UVCDspectrumofaqMutL-CTDwasobtainedinthe presence of Zn2+ to further characterize the Zn2+-bound state of Recent endonuclease assays have suggested that Mn2+ is aqMutL-CTD. No change was observed in the CD spectrum, essential to activate the endonuclease activity of aqMutL-CTD which revealed that Zn2+ binding did not perturb the secondary PLOSONE | www.plosone.org 5 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity Figure3.ThecharacterizationofinteractionsbetweenaqMutL-CTDandZn2+.(A)The1H-15NHSQCspectraof100mMaqMutL-CTDatthe variousconcentrationsofZn2+aresuperimposed.Therepresentativepeaks(C402andG405)showingtheslowexchangeareindicatedwiththeone- letteraminoacidcodeandresiduenumber.(B)ThechangeinthechemicalshiftofC402andG405.(C)PutativebindingsitesofZn2+aremagnified (seetheMaterialsandMethods).ThesidechainsofH353,E357,C371,C402,andH404areshown.ThebluespheresindicateZn2+ions. doi:10.1371/journal.pone.0098554.g003 invitro[12].Otherstudieshavealsoclarifiedthathumanandyeast together with information on the characteristic coordination MutLashareadependence onMn2+endonucleaseactivity[8,9]. geometry of Mn2+ in the Mn2+-bound protein [30], we coordi- NMR-based analysis was performed to examine the Mn2+ nated Mn2+ around E357, C402, and H404 to visualize Mn2+ binding site of aqMutL-CTD at the residue level in solution. On binding sites (Figure4D). thebasisoftheeffectoftheparamagneticrelaxationenhancement Far-UV CD measurements were performed to investigate the (PRE) of Mn2+, we expected to detect even transient and weak effectsofMn2+bindingonchangesinthesecondarystructureand intermolecular interactions by observing a decrease in the peak thermal stability (Figure 2). Although the far-UV CD spectrum intensitybecausePREwaspreviouslyshowntobeproportionalto was almost identical to that in the absence of metal ions the inverse sixth power of distance [29]. Although no significant (Figure 2A), the thermal transition curve was slightly different perturbationwasdetectedinthechemicalorpseudo-contactshift (Figure 2B). Mn2+ binding shifted the transition curve to slightly (Figure4A),markeddecreaseswereobservedintheNMRsignals higher temperatures. These results implied that Mn2+ interacted ofCTDwiththeadditionofMn2+(Figure4B).Overalldecreases less specifically and strongly with CTD than with Zn2+, which inthepeakintensityimpliedthatthewholemoleculeofaqMutL- were consistent withtheNMRresults. CTD was susceptible totheeffects ofthesolventPRE of Mn2+. Large decreases in intensities were observed in several InvestigationofIntermolecularInteractionsbetweenATP characteristic regions (peak intensity ratio ,0.52): the region of and aqMutL-CTD a1includingtheDQHA(X) E(X) motif,thea1–a2loop,andthe AlthoughATPbindingtoMutLiskeytoMMR,intermolecular 2 4 a2-b4 loop including the CPHGRPI motif (Figures 4B and 4C). interactionsbetweentheCTDandATPremainlargelyunknown Among them, a marked decrease in the peak intensity ratio (, [11,12,14,15,31,32]. Therefore, we obtained residue-based infor- 0.22) was observed in L354, E357, C402, and G405. Taken mationonATPbindingusingNMRspectroscopy.Thetitrationof PLOSONE | www.plosone.org 6 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity Figure4.Characterizationof interactionsbetweenaqMutL-CTDandMn2+.(A)The1H-15NHSQCspectraof100mMCTDatthevarious concentrationsofMn2+aresuperimposed.(B)ThepeakintensityratioofCTDintheabsenceandpresenceof400mMMn2+wasplottedagainstthe residuenumbers.Theuppersolidandmiddledashedlinesrepresent0meanvalue(0.52)0and0meanvalue2standarddeviation(0.37)0.Thebottom dottedlinesignifies0meanvalue226standarddeviation(0.22)0.(C)Mn2+bindingsitesbasedonthepeakintensityratioinBareshownwiththe colorcode:red,peakintensityratio,0.22;yellow,0.22,peakintensityratio,0.37.(D)ThebindingsitesforMn2+aremagnified.Thepurplesphere indicatesMn2+. doi:10.1371/journal.pone.0098554.g004 ATP against the CTD showed the shift of peaks with a fast chemical shift can be perturbed by the changes in electronic exchange regime, which revealed the intermolecular interactions environments around NMR-active nuclei of interest. Therefore, between the two molecules with weak affinity (Figure 5A). NMR severalfactorsinthebindingsysteminduceperturbationsofNMR PLOSONE | www.plosone.org 7 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity chemical shifts such as the ligand binding without the conforma- ATP shares a binding site with DNA, endonuclease activity of tionalchange and/ortheligand bindingwiththeconformational CTD should beaffected by thepresenceof ATP. change. It is often difficult todistinguish the mixed factors which However,therecentstudybyShimadaetal.demonstratedthat are attributed to the apparent change in the chemical shift using Mn2+-dependent endonuclease activity of Thermas thermophilus onlyNMRdata.Therefore, althoughwecannotasserttheorigin MutL (ttMutL)-CTD did not depend on the presence of ATP. ofalotofchemicalshiftchangesofaqMutL-CTD,ATPobviously Endonuclease activity of ttMutL-CTD in the absence and binds to aqMutL-CTD and ATP-binding may induce structural presence showed same activity [13]. Moreover, Mn2+-dependent rearrangements. endonucleaseactivityofaqMutL-CTDwasalsonotstimulatedor Thedetailedanalysisfortheaffinitybasedonthechemicalshift inhibitedbyAMPPNP(FukuiK.,unpublisheddata).Theseresults perturbation was hampered by the protein aggregation in the indicate that ATP does not competitively bind to the dsDNA presenceofthehighATP-concentrationover12mM.Itshouldbe binding siteof MutL-CTD withDNA,andvice versa. noted that examining of CTD aggregation is important to Taken alltogether, thebinding siteonaqMutL-CTD forATP understand endonuclease activity of MutL-CTD. Therefore, we which was revealed in this study is novel and is not used for the compared the intensity of CTD at each ATP concentration to DNA binding(i.e., an active siteof thenuclease). examine the formation of aggregates. The results showed the comparable intensities: the NMR peak intensity of CTD in the Investigation of ATP Binding to aqMutL-CTD presence of ATP with the 100-fold higher concentration than In order to obtain more information on binding specificity of CTD did not decrease compared to the peak intensity of CTD ATPtoaqMutL-CTD,wefurthercarriedouttheNMRtitration without ATP, indicative of the absence of aggregation. This is experimentsonADPbindingtoaqMutL-CTD(FigureS7)andthe because, in general, the formation of aggregates decreases ITC measurements on both ATP and ADP binding to CTD, significantly the intensity of the NMR signals. In order to obtain respectively (Figure S8). more evidence, we additionally performed the light scattering A series of the NMR titration spectra of aqMutL-CTD with measurements of CTD in the absence and presence of ATP ADP revealed the large differences from those with ATP. (FigureS9).Highandlowintensityoflightscatteringindicatelarge Although the ADP titration spectra also showed the changes of and insoluble aggregates as well as small and soluble proteins, NMRpeaks,thenumberofperturbedpeaksforADPtitrationwas respectively. For the reference intensity of aggregates, we also lessthanthatforATPtitration.Interestingly,allofthetrajectories carried outthe light scattering measurement of the amyloid fibril of the shifted peaks upon ADP titration including the peaks (e.g. of Ab peptides which is an ordered aggregate. As shown in 1–40 G405,Y409,andG422)whichshowedthetwodifferentdirections Figure S9,theintensitiesof CTD intheabsenceand presenceof of the trajectory of the peak shift upon ATP titration were linear ATPweremuchlowerthanthoseoftheamyloidfibrilsofAb 1–40 (Figure S7A). In addition, the magnitude of chemical shift peptides and were similar to the intensity of monomeric Ab 1–40 perturbation on ADP titration was lower than that on ATP peptides.Allofthesedataindicatedtheabsenceofaggregationof titration (Figure S7B). These results indicate that ADP also binds CTD in the presence of excess amounts of ATP but less than to aqMutL-CTD (Figure S7C) with the lower specificity and 12mM. binding affinity than that forATPbinding. Chemical shift perturbation analysis also revealed the binding At the same time, we can also exclude the possibility that sitesforATP.Theresiduesina1(K365),theloopbetweena1and perturbationofpeaksbyATPandADPtitrationwasnotinduced a2 (I373 and S374), and the C-terminal parts (G405, I408, and bytheionicstrengthchangeofNMRsolventbyadditionofATP V421)showedahighchemicalshiftperturbation(CSP)over0.082 and ADP. If there is an influence from the ionic strength of the onATPbinding(Figure5B).Theperturbedresiduesweremapped on the model structure depending on the CSP values using the addednucleotides,theshiftofNMRpeaksofaqMutL-CTDupon color codes (Figure 5C). additionofATPorADPshouldshowasimilarpattern.However, as described above, the pattern of trajectory between ATP and The docking simulation, which incorporated the CSP data, provided a putative scheme for the intermolecular interactions ADPtitrationwasobviouslydistinct,indicativeofspecificbinding between the CTD and ATP. The results obtained indicated that of ATPandADPtoaqMutL-CTD. ATP was bound to the cleft formed by a1 including the ITCresultsgaveafurtherinsightintoATPandADPbinding. DQHA(X) E(X) E motif, the a1–a2 loop including the ACRISV The calorimetric titration showed endothermic reaction heat for 2 4 motif, and the a2–b4 loop including the CPHGRPI motif titrating ATP or ADP to aqMutL-CTD (Figure S8). The (Figure5D).ThesidechainsofR358,K365,andN398interacted magnitude of observed heat for ATP titration was greater than electrostatically with the b-, a-, and c-phosphates of ATP. The thatforADPtitrationbyapproximatelyfive-fold,indicatingmore riboseandadenineofATPwereboundprimarilybeneaththea2– tightintermolecularinteractionsbetweenATPandaqMutL-CTD b4loop. than those between ADP and CTD. Both titration was not Thefar-UVCDspectrumandthermaltransitioncurveofCTD saturatedatthemolarratio([nucleotide]/[aqMutL-CTD])ofeven in the presence of ATP were identical to those without ATP, 25,consistentwiththeresultsofNMRtitration.Mostimportantly, indicating that weak ATP binding did not change the secondary theendothermic peaks forbothtitration decreased graduallyand structure andglobal thermal stability(Figure 2). thepeaksofATPtitrationweremorerapidlydecreasedthanthose Finally, to rule out the possibility of sharing binding sites for forADPtitration.Althoughthebindingcurvesweretoogentleto DNA, i.e., competitive binding between DNA and ATP, we analyze the data precisely, these ITC data suggested that ATP directly measured the binding of DNA to aqMutL-CTD. As binding is more specific than ADP binding with high binding showninFigureS6,excessamountsofDNArelativetoaqMutL- affinity compared to ADP binding. The different binding affinity CTD did not give any changes in the NMR signals of aqMutL- andspecificityofATPandADPtoMutL-CTDmaycomefroma CTD,indicatingthatbindingaffinityofDNAtoaqMutL-CTDis phosphate. extremely low. Although a binding region for DNA was not ItshouldbenotedthatfurtherbindingassaysusingGTP,NTP, determined experimentally, it is obvious that binding affinity of ordATParerequiredtoobtainaninsightintomolecularoriginon aqMutL-CTDforATPishigherthanthatforDNA.Asaresult,if binding specificity ofATPtoMutL-CTD. PLOSONE | www.plosone.org 8 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity Figure 5. The characterization of interactions betweenaqMutL-CTDand ATP. (A) The 1H-15N HSQC spectra of 100mM CTD at various concentrationsofATParesuperimposed.Theresiduesdisplayingsignificantperturbationsarelabeledusingaone-letteraminoacidcodeandresidue numbers.PeaksofC402(?)disappearedafter[ATP]/[CTD]=80asthetitrationproceeded;itwasthenruledoutinthechemicalshiftperturbation (CSP)analysis.However,thelabelwasdisplayedforthediscussionbelow.(B)CSPsoftheCTDinthepresenceof8mMATPwereplottedagainst residuenumbers.Thebottomsolidlinerepresentsthe0meanvalue(0.045)0.Themiddleandupperdottedlinessignify0meanvalue+0.56standard deviation(0.064)0and0meanvalue+standarddeviation(0.082)0forsignificantchanges,respectively.(C)Thedegreeoftheperturbationwasmapped PLOSONE | www.plosone.org 9 June2014 | Volume 9 | Issue 6 | e98554 ATP-DependentResponseofEndonucleaseActivity ontothemodelstructurewiththecolorcode:Red,CSP.0.082,yellow,0.082.CSP.0.064.(D)ThecomplexstructureoftheCTDandATPasobtained bythedockingsimulationusingMyPresto/Sievgene(seetheMaterialsandMethods).Theinteractingsitesareenlarged(right).ATPandaqMutL-CTD arerepresentedbyaballandstickmodelandmeshedsurfacemodel,respectively. doi:10.1371/journal.pone.0098554.g005 Discussion (E357K and H404A) displayed the inactivation of the Mn2+- dependentincisionofsupercoiledDNA,whichmayhavebeendue We here demonstrated that Zn2+, Mn2+, and ATP bound to todecreasesinthebindingcapabilityoftheCTDforMn2+[41]. aqMutL-CTD without alternating the secondary structure. Only These findings support the results of the present study in which Zn2+ markedly increased thermal stability, and this has been E357 and H404 consisted of binding sites for Mn2+, as shown in attributed to changes in tertiary structures and/or dynamics. It Figure 4B. should be noted that although the complete profile of thermal Mn2+ shared CTD binding sites with Zn2+, with similar unfolding of aqMutL-CTD in the presence of Zn2+ was not coordinationgeometries(Figures3and4).BecauseZn2+enhanced obtained, it is obvious that cooperativity of thermal unfolding of theendonucleaseactivityofbsMutLinthepresenceofMn2+[1], Zn2+-bound aqMutL-CTD is higher than that of apo-CTD, these two divalent metal ions may be coordinated together for indicating the increase in thermal stability. Many new NMR MMR invivo to ensure both stability and function; however, signals can be interpreted by reorganizing large parts of the detailed future work is required to clarify these intermolecular tertiary structure (Figure 3A). Several key residues suchas H353, interactions. E357, C371, C402, and H404 may be important for suppressing ATP binding to MutL and MutLa for MMR has been global motion due to the tight binding of Zn2+ (Figure 3B). demonstrated and the NTD was predominantly suggested to PreviousstudiesalsoreportedthatZn2+ionsservedasastructural accommodate ATP [14,31]. However, there has been little focus factor rather than a catalytic contribution [33–35], implying a on ATP binding to the CTD and no consensus on their relationship between Zn2+ binding and stability. Since aqMutL- intermolecular interactions. Mauris J. et al. showed that ATP did CTD functions at high temperatures above 90uC, Zn2+ binding not bind aqMutL-CTD using the filter binding assay and further shouldbeessentialformaintainingconformationalstabilityforits indicatedtheabsenceofATPbindingmotifinaqMutL-CTD[41]. function. Incontrast,itcouldbeimpliedthattheintermolecularinteractions Mn2+hasbeenshowntoplayapivotalroleintheendonuclease betweenCTDfromttMutLandATPmaybepossible[42].This activityofaqMutL-CTD.Althoughithasbeensuggestedthatthe discrepancy may reflect the difficulty in detecting weak intermo- coordinationnumberandresiduecompositionoftheZn2+-binding lecular interactions, which are markedly susceptible to subtle site of MutL may preclude Mn2+ binding based on the changes in the surrounding conditions used such as temperature, bioinformatics of metal-binding proteins [35–40], we clarified salt concentration, andpH. thattheMn2+bindingsitesofaqMutL-CTDwerelocatedaround We here demonstrated weak and specific interactions between the endonuclease activity sites between the two motifs, ATP andaqMutL-CTD using solution-state NMR at the residue DQHA(X) E(X) E and CPHGRPI (Figure 4). The reported level and ITC at the molecular level. By incrementally adding 2 4 endonuclease activity assay of the aqMutL-CTD mutants excess ATP to aqMutL-CTD, we revealed gradual shifts in peak positions, which indicated weak intermolecular interactions. The linearshiftfollowedsuccessivelybyalinearshiftinadifferentway, as represented by G405, I408, and Y409 (Figure 5), implied the two states of the ATP-bound CTD. Considering the concentra- tionsofATPandCTDusedhere,theaffinityoftheCTDforATP was markedly lower than that of the NTD, the dissociation constantofwhichisapproximately20nM(FukuiK.,unpublished data). ITC data also showed the weak affinity of CTD for ATP, consistent withtheNMRresults. Chemicalshiftperturbationanalysisandthedockingsimulation allowedustodeterminethebindingsitesforATP,whichincluded themotifsconservedinthevariousspecies(FigureS5),andimplied that the binding of ATP to the CTD may be a common feature. The feature of the rigid conformations of the a2–b4 loop containingtheconservedCPHGRPImotifduetohydrogenbond networks (Figure 1D) may be involved in the effective binding of Figure6.Activation-attenuationmodelfortheATPconcentra- metal ions and ATP. Interestingly, the ATP binding sites of tion-dependent response of MutL endonuclease activity. The aqMutL-CTD were included in the regions responsible for the threeconformationalstatesofMutL(rest,active,andinhibitedstates) interdomaininteractionsfortheNTD,whichhavebeensuggested are in equilibrium. The dominantconformational state can be shifted topromotetheendonucleaseactivityoftheCTD[12,13,15].This depending on the differences in the relative concentration between indicatesthepossibilityofregulatoryroleofATPbindingtoCTD MutL and ATP. The rest state of MutL in the absence of ATP (left) is predominantlypopulated.ThesmallincreaseintheATPconcentration inCTD-NTD interdomain interaction. drives the ATP binding to NTD thereby stabilizing the active state In addition, Fukui K. and coworkers modified Cys496 of (middle).AtthehighATPconcentration, however,theinhibitedstate ttMutL-CTD, which is the conserved cysteinyl residue in whichsuppressesendonucleaseactivityisfavoredbyaccommodating CPHGRPI motif corresponding to Cys402 of aqMutL-CTD further ATP to CTD (right). It should be noted that the complicated [42]. They showed that chemical modification of Cys496 with conformationalstatesofATP-boundCTDwhichareATPconcentration- dependent are not considered for simplicity. The ellipses and circles DTNB and the substitution of Cys496 with an Ala residue in indicatetheCTDandNTD,respectively.BluerectanglesshowATP. ttMutL caused the decrease in the efficiency of AMPPNP- doi:10.1371/journal.pone.0098554.g006 dependent suppression of endonuclease activity. Consideringthat PLOSONE | www.plosone.org 10 June2014 | Volume 9 | Issue 6 | e98554