4384ndash4395 Nucleic Acids Research 2007 Vol 35 No 13 Published online 18 June 2007doi101093nargkm448

Conformational and thermodynamic changesof the repressorDNA operator complex uponmonomerization shed new light on regulationmechanisms of bacterial resistance againstb-lactam antibioticsJulien Boudet1 Valerie Duval2 Helene Van Melckebeke1 Martin Blackledge1

Ana Amoroso23 Bernard Joris2 and Jean-Pierre Simorre1

1Institut de Biologie Structurale Jean-Pierre Ebel CEA-CNRS-UJF 41 Avenue Jules Horowitz 38027 GrenobleCedex 1 France 2Centre drsquoIngenierie des Proteines Institut de Chimie B6A Universite de Liege Sart-TilmanB4000 Belgium and 3Catedra de Microbiologıa Facultad de Farmacia y Bioquımica Universidad de Buenos AiresJunın 954 (1113) Buenos Aires Argentina

Received April 11 2007 Revised May 16 2007 Accepted May 18 2007

ABSTRACT

In absence of b-lactam antibiotics BlaI and MecIhomodimeric repressors negatively control theexpression of genes involved in b-lactam resistancein Bacillus licheniformis and in Staphylococcusaureus Subsequently to b-lactam presence BlaIMecI is inactivated by a single-point proteolysis thatseparates its N-terminal DNA-binding domain to itsC-terminal domain responsible for its dimerizationConcomitantly to this proteolysis the truncatedrepressor acquires a low affinity for its DNA targetthat explains the expression of the structural genefor resistance To understand the loss of the highDNA affinity of the truncated repressor we havedetermined the different dissociation constantsof the system and solved the solution structureof the B licheniformis monomeric repressor com-plexed to the semi-operating sequence OP1 of blaP(12OP1blaP) by using a de novo docking approachbased on inter-molecular nuclear Overhausereffects and chemical-shift differences measuredon each macromolecular partner Although theN-terminal domain of the repressor is not subjectto internal structural rearrangements uponDNA binding the molecules adopt a tertiaryconformation different from the crystallographicoperatorndashrepressor dimer complex leading to a308 rotation of the monomer with respect to acentral axis extended across the DNA

These results open new insights for the repressionand induction mechanisms of bacterial resistance tob-lactams

INTRODUCTION

The Bacillus licheniformis BlaI protein (BLBlaI 128 aminoacids) is a transcriptional repressor of the BlaPb-lactamase This enzyme is a specific hydrolase ofb-lactam antibiotics induced in response to the presenceof this class of antibiotic outside the cell (1) BLBlaI ishomologous to Staphylococcus aureus BlaI (SABlaI 126amino acids) and MecI (SAMecI 123 amino acids)regulators involved in the induction of BlaZ b-lactamaseand resistant penicillin-binding protein 2a (SAPBP2a orSAMecA) respectively (2) BlaZ b-lactamase and resistantPBP2a are the main factors involved in staphylococcalb-lactam antibiotic resistance The BlaIMecI repressorsare organized in two domains an N-terminal domain(NTD) for DNA binding and a C-terminal domain (CTD)for repressor dimerization (3) The crystallographic 3Dstructures of SABlaI and SAMecI dimers in free and incomplex with their DNA operators have been determined(4ndash6) and the 3D structure of BLBlaI N-terminal DNA-binding domain has been determined by heteronuclearnuclear magnetic resonance (NMR) spectroscopy (7) Forthe three repressors the BlaIMecI-NTD share a commonfold composed of three a-helices and three b-strandstypical of the winged helix regulator proteins The SABlaISAMecI structures highlight dimers of two independentN-terminal DNA-binding domains and two intertwined

To whom correspondence should be addressed Tel +33-4-38785799 Fax +33-4-38785494 Email jean-pierresimorreibsfr

2007 The Author(s)

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (httpcreativecommonsorglicenses

by-nc20uk) which permits unrestricted non-commercial use distribution and reproduction in any medium provided the original work is properly cited

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

C-terminal dimerization domains (Figure 1A)The BlaIMecI repressors bind specifically to similarnucleic sequences composed of an imperfect dyadsymmetry (24 to 30 bp) containing a central conservedpalindrome 50-TACANNTGTA-30 (8) (nucleotide one-letter code N for any nucleotide) The study of BLBlaIdimerization and its interaction with its bla operator(BLOPbla) has shown that at a concentration below thedissociation constant of BLBlaI dimer ([BLBlaI]525 mM)the binding of one BLBlaI monomer to its operator leadsto the binding of the second monomer with an infinitecooperativity (9) In this way it has proved impossible toisolate a BLBlaI monomer bound to its DNA operator(Figure 1B) In addition for the same repressor it hasbeen shown that the BLBlaI-NTD obtained by papainproteolysis retains its capacity to bind BLOPblaHowever its affinity for its DNA target becomes at least500 to 1000 times lower as determined by DNAseIfootprinting experiments (3)

In S aureus and B licheniformis the genes encoding forb-lactam resistance (blaZ blaP and mecA) form adivergon with the blaIblaR1 and mecImecR1 operonsrespectively (1011) blaR1mecR1 encodes a penicillinreceptor essential for the induction of the gene ofresistance The blamec operators are located in theintergenic DNA sequence between the blaPblaZmecAgene and the blaImecI-blaR1mecR1 operon In presenceof b-lactam antibiotics the BlaR1MecR1 receptor isacylated The resulting activated receptor launches acytoplasmic signal which inactivates BlaIMecI repressor

In S aureus the SABlaI inactivation is achieved by theproteolysis of the peptide bond linking residues 101 and102 giving rise to SABlaI-NTD dissociated from SABlaI-CTD (12) The truncated SABlaI-NTD is a monomerpresents a lower affinity for its DNA operator and isreleased into the cytoplasm For BLBlaI repressor the

presence of a coactivator (13) generated by the activation ofthe BlaR1 repressor has been postulated The binding ofthe coactivator to BLBlaI would then result in a decreasedaffinity of BLBlaI repressor for its DNA target (14)To better understand the mechanisms of the binding

of the BlaIMecI repressors on the DNA and theirinactivation during the induction of BlaPBlaZMecAproteins we solved the solution structure of the lowaffinity BLBlaI-NTD12OP1blaP complex ([BLBlaI-NTD][12OP1blaP]) The dissociation constants of theSAMecI in complex with the B licheniformis operator OP1

of blaP (OP1blaP) and in complex with the S aureusoperators of blaZ (OPblaZ) and mecA (OPmecA) havebeen determined NMR spectroscopy allowed us toestimate the dissociation constants of the SAMecI-NTDand BLBlaI-NTD for the semi-operating sequence of theB licheniformis blaP gene OP1 (12OP1blaP) and for thesemi-operating sequence of the S aureus mecA gene(12OPmecA)

MATERIALS AND METHODS

Sample preparation

The pET22b was used as vector for the overproductionof the His-tagged SAMecI protein (SAMecI-His6)The SAMecI coding sequence was amplified by PCRfrom the S aureus ATCC 43300 genomic DNA usingTaq polymerase (Promega) and the following oligonucleo-tides as primers 50-GAG-CAT-ATG-GAT-AAT-AAA-ACG-TAT-GAA-ATA-TCA-TC-30 and 50-CTC-GAG-TTT-ATT-CAA-TAT-ATT-TCT-CAA-TTC-TTC-TA-30

purchased from EUROGENTEC SA Belgium (httpeurogenteccom) The fragment generated correspondsto the SAMecI coding sequence within the restrictionsites for NdeI and XhoI and was cloned into the pCR4TOPO (Invitrogen) vector to generate the pCR4-mecI

Figure 1 Dimeric interaction of the BlaIMecI repressor with its DNA operator (A) Representation of the S aureus dimeric MecI repressor (ribbon)in interaction with the bla operator (PBD ID 1SAX) C-terminus of MecI (dimerisation domain) is colored in pink and N-terminus (DNA bindingdomain) in blue (B) Representation of the two potential parallels pathways represented for the B licheniformis BlaI repressor with the bla operatingsequence

Nucleic Acids Research 2007 Vol 35 No 13 4385

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

The identity of the sequence was verified before thepCR4-mecI was digested with NdeI and XhoI enzymes andcloned into the pET22b to generate the pCIP451 whichcontains the MecI coding sequence with a polyhistidinetag at its carboxy-terminal endSAMecI-His6 samples were overexpressed in E coli

strain BL21(DE3) For the production of uniformly15N-labeled samples Luria-Bertani medium was replacedby a M9 minimal medium supplemented with 11 gl15NH4Cl 2mM MgSO4 01mM CaCl2 and 2 gl glucoseCells were grown at 378C to a 600 nm absorbance of 06and 05mM IPTG was added for a 3-h induction periodCells were then harvested by centrifugation re-suspendedin buffer A (20mM NaH2PO4Na2HPO4 buffer 500mMNaCl pH 76) and disrupted by passage through anInceltech disintegrator The soluble fraction was separatedby centrifugation at 40 000 g and loaded onto a NiPDCchelating column (26 10 cm2 Affliland) charged with50mM NiSO4 and equilibrated with buffer A TheSAMecI-His6 protein was eluted by a gradient of bufferB (250mM imidazole 500mM NaCl pH 8) The proteinwas dialyzed against buffer A and concentrated The finalyields of labeled and unlabeled purified protein wererespectively 9 and 15mgl of cell culture The isotopiclabeling of 95 was determined by mass spectrometryThe uniformly 13C15N-labeled BLBlaI sample was

prepared as described in Van Melckebeke et al (7) TheSAMecI-NTD protein was dialyzed into a 75mMNaH2PO4Na2HPO4 buffer 200mM KCl 1mM EDTA1mM NaN3 pH 76 and was concentrated to 05mM byultrafiltration through a 5 kDa cut-off Amicon for furtherNMR analysisUnlabeled single-stranded DNA samples of NMR

quality were chemically synthesized and purified byEUROGENTEC SA Belgium (httpeurogenteccom)Freeze-dried samples were suspended in the buffers usedfor interaction studies Single-stranded DNA were mixedin a 11 ratio subsequently heated to 1008C and slowlycooled down at room temperature in order to improveintermolecular arrangements The mecA semi-operatorsequence [50-ATA-AGA-CTA-CAT-30 and complemen-tary strand 50-ATG-TAG-TCT-TAT-30] was designedaccording to previous papers results (7) and obtained ata 9mM final concentration in 75mM NaH2PO4Na2HPO4 buffer 200mM KCl 1mM EDTA 1mMNaN3 pH 76 The OP1blaP half-dyad [50-AAA-GTA-TTA-CAT-30 and 50-ATG-TAA-TAC-TTT-30] wasselected using former interaction results with BLBlaI-NTD and obtained at a 23mM final concentration

NMR spectroscopy

NMR experiments were performed on Varian Inova 600and Inova 800 spectrometers both equipped with a triple-resonance (1H 13C 15N) probe and shielded z-gradientsFurthermore the Varian Inova 800MHz spectrometer isequipped with a cooled probe The temperature was set to298K Proton chemical shifts were referenced with respectto an external DSS calibration 13C and 15N chemicalshifts were accordingly referenced indirectly using the1HX following ratios 0251449530 (13C) and 0101329118

(15N) All experiments used the pulse sequences providedby the Varian Protein Pack (httpvarianinccom) Dataprocessing and peak intensity measurements were per-formed using the NMRPipe program Peak picking andspectra display were achieved using the NMRViewsoftware

Affinity measurements

Electrophoretic mobility shift assays were carried outusing an ALFexpress DNA sequencer as described in theliterature (11) The CY5-labeled fluorescent double-stranded oligonucleotides used in these experiments arelisted in Table 1 (1516)

For NMR interaction studies 1H-15N HSQC spectrawere collected along the titration of SAMecI and theBLBlaI truncated repressors with the 12 bp DNA To limitdilution and favor sensitivity low volumes of highlyconcentrated half-operators of mecA and blaP genes wereadded to each protein sample 15N-labeled SAMecI-NTDand BLBlaI-NTD samples concentrations were set to01mM Titration experiments led on 15N-labeled MecI-NTD (respectively 15N-labeled BLBlaI-NTD) with bothunlabeled mecA and blaP semi-operating sequences wereperformed in a 75mM (respectively 50mM) NaH2PO4Na2HPO4 buffer with 200mM KCl 1mM EDTA 1mMNaN3 and 10 of D2O at a controlled pH of 76

Data analysis and Kd calculation were performed withthe titration script developed for the NMRView software(17) In each case curves fitting were displayed by theXmgrace software (httpplasma-gateweizmannacilGrace)

NMR structural restraints

1H 13C and 15N assignment of the free BLBlaI-NTDprotein has been previously reported and deposited in theBMRB (accession number 5873) Resonance assignmentof the unlabeled 12 bp DNA of the OP1blaP semi-operating sequence was performed on a 1mM sample in50mM of NaH2PO4Na2HPO4 200mM KCl 1mMEDTA 1mM NaN3 pH 76 in 9010 H2OD2O ATOCSY spectrum with 80ms mixing time and a NOESYspectrum with 150ms mixing time were recorded on thatsample 1H-1H NOESY was also collected in 100 D2Owith 150ms mixing time A classical homonuclear DNA

Table 1 Oligonucleotides used in band-shift assays

Operators Oligonucleotide sequences

B lichenOP1blaP

50CY5-AAA GTA TTA CAT ATG TAA GAT TTA-30

30-TTT CAT AAT GTA TAC TTA CTA AAT-50

S aureusOPblaZ

50CY5-TAA AAA TTA CAA CTG TAA TAT CGG-30

30-ATT TTT AAT GTT GAC ATT ATA GCC-50

S aureusOPmecA

50CY5-ATA AGA CTA CAT TTG TAG TAT ATT-30

30-TAT TCT GAT GTA AAC ATC ATA TAA-50

The nucleotide sequences correspond to the blaP operator fromBlichen (12) the blaZ and the mecA operators form S aureusNumbering used in this study for B licheniformis BlaP start from the 50

end (A1 A2 A3 ) The complementary strand is noted with a star(T1 T2 T3 ) and numbering starts from 30 end

4386 Nucleic Acids Research 2007 Vol 35 No 13

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

assignment strategy was used (18) Due to the largenumber of overlap in H4

0 and H50H5

00 regions assignmentand chemical-shift mapping were restricted to H1

0 H20

H30 H5 and H6H8 resonances For the 12 bp 12OP1blaP

assignment in the complex we reiterated the procedureused for the free DNA but using only a 2D filteredNOESY experiment recorded on the [13C-15N BLBlaI-NTD][12OP1blaP] sample diluted in 100 2H2O Theexperiment was recorded on the 800MHz spectrometerwith a mixing time of 150ms

For the [BLBlaI-NTD][12OP1blaP] complex the1H 13C and 15N protein resonances were not re-assignedab initio Assignment of the protein in complex wasobtained by comparison of 1H-13C HSQC and a1H-15N HSQC of the free and bound formsExperiments were recorded with 2mM samples and a[13C-15N BLBlaI-NTD][12OP1blaP] molar ratio of 1 in100 D2O and in 90 H2O To confirm this assignmenta 3D HC(C)H-TOCSY experiment was also recorded onthe complex The weak dependence of 13C chemical shiftsto long distances variation and then to the complexformation has permitted to verify each correspondingresidue assignment

In order to obtain inter-molecular NOE restraintsbetween the two partners of the [BLBlaI-NTD][12OP1blaP] complex we collected isotopically doublyfiltered 2D and 3D 13C15N NOESY HSQC with mixingtimes set to 150ms A protein chemical-shift mapping wasobtained by comparison of 1H-15N HSQC methyl-selective 13C HSQC and 13C HSQC optimized foraromatics recorded on the free and bound forms of theprotein A chemical-shift mapping for the blaP half-operator was obtained by comparison of the 1H-1HNOESY recorded on the free DNA and the 2D filteredNOESY experiment recorded on the 11 [13C-15N BLBlaI-NTD][12OP1blaP] sample

Docking procedure

Determination of the structure of molecular complexesfrom sparse NMR data is a difficult task requiring theintegration of local and long-range molecular plasticity(19) In order to allow the maximum available degrees offreedom we have determined the quaternary architectureof the BlaI repressorDNA operator complex using ade novo approach starting from randomized coordinatesExperimental NOE collected on the 15N-13C BLBlaI-NTDfree protein were used to fold the bound polypeptide (7)The use of these constraints is based on the observationthat few chemical shifts change between the bound andfree forms showing that the fold of the two proteins isessentially the same The 12 bp operator was constrainedusing distance restraints extracted from a canonicaldouble-strand DNA B-DNA standard angles (a b d ge z c u0 u1 u2) were also included for each nucleotideto facilitate the DNA helix fitting

Distances restraints extracted from intermolecular NOEwere defined at 5 A (one DNA proton correlated with oneprotein resonance) A thirdly set of structural data wasintroduced using chemical-shift mapping performed onthe two molecules Chemical-shift-derived distance

restraints were created by combining lsquosignificantrsquo chemi-cal-shifts variations identified on the polynucleotide1H-1H NOESY with shifted residues in 15N-HSQC and13C-methyl-selective-HSQC spectra and including theseconstraints in an ambiguous manner (2021) Distancesinferred from these distance restraints were fixed at 10 A inthe docking simulationAll calculations were performed using the program

Discover with the AMBER4 force field (22) Simulatedannealing was used to explore the conformationalspace for the structure determination and a restrainedmolecular dynamics calculation was used to refine eachstructure (23) Detailed initial conditions and physicalcharacteristics of exploratory period have been reportedpreviously (24)

RESULTS

Interaction between SAMecI and the DNA operators

The binding curves of SAMecI to its operator OPmecAand to the two b-lactamase operators OP1blaP andOPblaZ have been determined by band-shift assay Allshow sigmoidal-binding curves as previously described forthe interaction of BLBlaI repressor with its operators (9)So as for BLBlaI the binding parameters of SAMecIinteraction include two equilibria and only the globaldissociation constant Kd=Kd1 Kd2 can be obtainedwhere Kd1 and Kd2 are the dissociation constants ofSAMecI dimer and SAMecI dimer-operator complexrespectively (Table 2)

NMR titration of truncated monomeric NTD with cognateand crossed semi-operators

Titration of 15N-labeled BLBlaI or SAMecI truncatedrepressor with progressive amounts of unlabeled DNAhalf-dyads was performed by monitoring changes in1H-15N HSQC spectra Significant chemical-shift changesfor correlation peaks in the 1H-15N HSQC spectra wereobserved when the 12OP1blaP and the 12OPmecAwere added to protein samples [DNA][Protein] molarratios were varied from 0 to 10 for [BLBlaI-NTD][12OP1blaP] [BLBlaI-NTD][12OPmecA] [SAMecI-NTD][12OPmecA] and from 0 to 50 for [SAMecI-NTD][12OP1blaP] The chemical-shift changes observedupon complexes formation reached a plateau over 7[DNA][protein] molar ratios for [BLBlaI-NTD][12OP1blaP] [BLBlaI-NTD][12OPmecA] [SAMecI-NTD][12OPmecA] and a plateau over 30 for [SAMecI-NTD][12OP1blaP] (Figure 2) These results confirm theformation of stable intermolecular interactions with

Table 2 Equilibrium parameters of the MecI-operator interactions

Kd (1015M2) Kd2(M)

OPmecA 17 04 68 1011

OPblaZ 14 07 56 1011

OP1blaP 4 2 16 1010

The values of Kd2 (Kd=Kd1Kd2) were estimated by supposing thatthe value of Kd1=25 mM estimated for B licheniformis BlaI dimer isthe same for S aureus MecI

Nucleic Acids Research 2007 Vol 35 No 13 4387

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

saturating DNA quantities At any point during thetitration specific single correlation peaks were detectedsuggesting that the truncated monomeric repressorssemi-operators complexes are in fast exchange regarding theNMR time scale Only well-resolved correlation peakswith a chemical-shift cut-off equal or superior to 003 ppmwere used in the four experiments The affinity constantswere calculated from a non-linear fit of the significantchemical-shift variations versus [DNA][protein] ratiousing equation given by Morton et al (25) Titrationdata were analyzed assuming that the observed chemical-shift perturbation is a weighted average between the twoextreme values corresponding to the free (d=0) and thebound state (d=dmax) so that

max

frac14

Opfrac12 0thorn Ifrac12 0thornKd

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiOpfrac12 0thorn Ifrac12 0thornKd

24 Opfrac12 0

Ifrac12 0

q

2 Ifrac12 0

where [Op]0 and [I]0 are the total molar concentrations ofDNA operator and protein Statistical analysis usingMonte-Carlo simulations were used to evaluate theuncertainty of the fitted parameters

Semi-operating sequences of the blaP and mecA genestitration curves of selected individual amino acid residuesresulted in averaged binding constants of 190 50 mM(Kd1) 170 50 mM (Kd2) and 160 60 mM for the[BLBlaI-NTD][12OP1blaP] the [BLBlaI-NTD][12OPmecA] and the [SAMecI-NTD][12OPmecA] com-plexes respectively (Figure 2) For [SAMecI-NTD][12OP1blaP] the measured affinity constant reaches860 80 mM (Kd3)

NMR structural determination

Considering the low affinity constant measured pre-viously we decided to investigate the solution structureof the [BLBlaI-NTD][12OP1blaP] complex using asparse data approach Structure prediction using a dock-ing approach remains difficult because of the number and

Figure 2 Determination of dissociation constants using chemical shift titration obtained by NMR Weighted sum of the 1H15N chemical shiftvariation measured for the protein on 15N-HSQC is plotted as a function of the DNAprotein ratio for the different complexes (A) B licheniformisBlaI-NTD with the B licheniformis 12-operator of the blaP gene and (B) with the S aureus 12-operator of the mecA gene (C) S aureus MecI-NTDwith the 12-operator of the mecA gene and (D) with the B licheniformis 12-operator of the blaP gene By fitting the curves the different dissociationconstants Kd were obtained Kd1 [BLBlaI-NTD][12OP1blaP]=190 50 mM Kd2 [BLBlaI-NTD][12OPmecA]=170 50 mM Kd3 [SAMecI-NTD][12OP1blaP]=860 80 mM and Kd4 [SAMecI-NTD][12OPmecA]=160 60 mM

4388 Nucleic Acids Research 2007 Vol 35 No 13

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

the variety of parameters that should be taken intoaccount Recent advances in the field (21) take advantageof structural restraints from experimental interaction data(biochemical andor biophysical) to determine the relativeposition of the molecular partners Such approaches havealso been applied to the determination of the quaternarystructure of proteinDNA complexes (26) Our de novodocking protocol has been performed as follows

To investigate the structure of the complex between the13C15N-labeled BLBlaI protein with the unlabeled blaPsemi-operating sequence we collected two filteredNOESY experiments The 2D isotopically filteredNOESY performed in 2H2O allowed us to observe threeNOE To verify these constraints we collected anadditional 3D NOESY 13C HSQC experiment in H2Oresulting in one additional inter-molecular contact Thesefour correlations have been assigned ambiguously with atolerance value of 005 ppm for the protein and 003 ppmfor the DNA

To complement the four NOE distance constraints achemical-shift mapping has been carried out for bothmacromolecular partners as follows For the protein wecompared 1HN and 15N (backbone amide) and 1HC and13C methyl chemical shifts assigned for the free and boundform of the molecules measured in 1HN-15N and 1HC-13Cmethyl-selective and aromatic-selective HSQC respec-tively Indeed methyl group and aromatic chemical-shifts perturbations were judged to be good additionalprobes to precisely localize the interaction siteConcerning the nucleic acids only the sugar H1

0 H20

H200 H3

0 and the base H6H8 protons were used forchemical-shift mapping due to the overlap problemsobserved in the other spectral regions

The chemical-shifts variations resulting from complexformation allowed us to make an inventory of eachindividual amino acid or nucleotide potentially involved inthe interaction We only considered chemical variations

larger than 08 ppm for the weighted sum of the 1H and13C or 1H and 15N chemical-shift variations of the proteinwith respect to the gyro-magnetic ratios and larger than009 ppm for the proton chemical-shift variation measuredfor the blaP half operator In both cases peaks thatdisappeared were incorporated in the docking procedureFor the blaPDNA operator the majority of perturbationsconcerned nucleotides in the vicinity of the T8ACA11A8TGT11 motif namely the thymines 5 7 8 and 6 911 on the complementary strand the adenines 6 9 7 8

and 12 and the guanines 4 and 10 (nucleotidesnomenclature is described in Table 1 legend) For theprotein larger perturbations are observed for the N-terminal part the H2 and H3 helix and around the wingThe structure calculation started from random coordi-

nates of the entire system and used 1513 experimentalintra-molecular NOE of the free form BLBlaI-NTDprotein and simulated distance restraints for a standardB-helix (613 intra-residue distances and 52 inter-residuedistances) Classical dihedral angles for DNA subunitswere also included namely 22 a 24 b 23 g 23 d 22 e 22z 24 24 u2 24 u1 and 24 u0 The use of experimentalNOE from the free form of the molecule is based on thelack of drastic shift between 1H-15N HSQC of theunbound and the bound protein indicating that nosignificant structural rearrangement occurs NMR dihe-dral angles were also calculated using TALOS softwareand incorporated in the calculation Four intermolecularNOE were associated with 11 and 24 additional ambig-uous distance restraints proceeding from 13C methyl-selective HSQC experiments and from 15N HSQC spectrarespectively

Structure of the complex

Ten lowest energy structures from 250 calculations of the[15N-13C BLBlaI-NTD][12OP1blaP] have been generatedusing a de novo driven docking (Figure 3) As expected the

Figure 3 Ten lowest energy structures obtained by de novo docking of the Bacillus licheniformis BlaI-NTD in interaction with the blaP semi-operating sequence (A) Front view and (B) back view of the structure ribbon representation Root mean square deviation has been calculated to07A for all heavy atoms of the 10 structures The 12 bp DNA of blaP half operator are shown in orange H3 helix is inserted into the major grooveand the minor groove is overhung by the wing motif

Nucleic Acids Research 2007 Vol 35 No 13 4389

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

global fold of the NTD repressor domain is not modifiedby the complex formation Pairwise RMSD calculated onthe NTD backbone atoms between the monomericcomplexed BLBlaI structure and free BLBlaI-NTDSAMecIOPblaZ and SABlaIOPblaZ is 1 16 and15 A respectively The typical structural arrangementsof the WHP family protein is conserved without violationsin the structure calculation file The three a-helices H1(9ndash20) H2 (26ndash36) H3 (41ndash54) and the three strandedb-sheets S1 (23ndash25) S2 (57ndash62) S3 (65ndash70) are packedfollowing the sequence H1ndashS1ndashH2ndashH3ndashW1ndashS2ndashS3 Thewing motif W1 consists of a short loop (residues 63 and64) Three-dimensional structures of the free and boundforms of the BLBlaI-NTD are very close except for themore dynamic residues located in the N- and C-terminalsextremities of the protein namely M1 to I4 and Y77 toS82 respectively Moreover it should be noticed thatrestricted conformational modification occurred for theresidues of the Wing (G63 and R64) and for few residuessurrounding this motif (E62 V65 and F66)In the complex the B-DNA form is conserved in the

final lowest energy structures The position of the bases isless well-defined at both ends of the molecule than in the

central part possibly due to a lack of conformationalrestraints Concaveness observed for the complete opera-tor for SABlaI and SAMecI is not observed for the halfoperator (Figure 4) However alignment of our 12 bpoperator with the common base of the full MecI operatorshows a very similar conformation (Figure 4B) Theabsence of a detectable kink could be due to the reducedlength of the half-operator used in the NMR study

To analyze the orientation of the protein relative to theDNA we have aligned our structure with the two X-raystructures of SABlaI and SAMecI using a superposition ofthe conserved DNA sequence Compared to the twocrystallographic structures the relative orientation of themonomeric protein with respect to the DNA helix shows a308 rotation along the long axis of the DNA and atranslation of 3 A (Figure 4B) To compare the stability ofthe different proteinDNA complexes analysis of theproteinDNA interaction using LIGPLOT software (27) ispresented in Table 3 Details of the nominative proteinDNA contacts observed in our structure are presented insupplementary Figure S1

Two helices (H2 and H3) the wing motif and theN-terminal domain of the B licheniformis BLBlaI-NTD

Figure 4 Front view of the superposition of the structure of the B licheniformis BlaI-NTD in interaction with the blaP semi-operating sequence andthe S aureus MecI dimer in complex with the blaP operator (1SAX) (A) Alignment of the two complexes using [BLBlaI-NTD][12OP1blaP] and oneof the [NTD][12OP] of SAMecISAOPmecA Secondary structures of the monomer BLBlaI-NTD conserved a similar position relative to the majorand minor groove of the DNA (B) Alignment of the two structures using only the blaP semi-operating sequence and the corresponding 12 bp DNAsequence of the SAOPmecA Monomer BLBlaI-NTD has twisted by 30 degrees relative to the sequence of the operator

Table 3 Number of proteinDNA interactions detected using LIGPLOT software for the different dimeric complexes S aureus SAMecIOPblaZ

(1SAX) S aureus SABlaIOPblaZ (1SXD) S aureus SAMecIOPmecA (2D45) and for our monomeric B licheniformis [BLBlaI-NTD][12OP1blaP]

Secondary structure elements SAMecI (1SAX) SAB1aI (1XSD) SAMecI (2D45) BLBlaI-NTD low ene BLBlaI-NTD Ensemble Average

N-TER 2 0 0 1 08H1 1 0 1 1 06H2 1 0 0 3 23H3 6 (3) 2 (1) 6 (3) 4 (3) 67 (52)WING 2 0 2 2 17OTHER 0 2 1 5 204 (04)TOTAL 12 (3) 4 (1) 10 (3) 16 (3) 145 (56)

Number of interactions has been obtained for the NMR lowest energy structure Average values are displayed for the 10 lowest energy structuralensemble Values in parentheses correspond to the number of sequence specific H-bonds involving atoms of the DNA bases

4390 Nucleic Acids Research 2007 Vol 35 No 13

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

have been reported to establish contacts with DNA In ourstructure the H3 helix (P41-K53) is deeply inserted in theDNA major groove whereas the minor groove is close tothe wing motif (G63-R64) For our NMR-based modelresidue R64 of the wing motif (G63-R64) binds to the30 end of thymines 3 and 2 Moreover interaction isoccasionally propagated for residues surrounding the wingmotif ie F66 and H61 which make contacts with A6 T7and C4 T3 nucleotides respectively In the MecIBlaIX-ray structures the wing amino-acid F67 binds the DNAbackbone in the opposite side to that observed in theNMR model (A7C and T8C equivalent to position G4and T5 in our DNA sequence) As a consequence of theglobal rotation contacts of the wing with the DNA arealso rotated with respect to the lsquocrystal wingrsquo position inSAMecIOPbla

The H3 helix residues (P41-K53) contact principallywith nucleotides belonging to the T8ACA11T8

GTA11

motif T43 T46 and R50 privilege interactions withnucleotides T9 G10 and T11 respectivelyFurthermore K54 (and W39) anchors the position ofH3 via interaction with A12 (and G10) In this lowaffinity complex Q45 plays a central role forminghydrogen bonds network with nucleotides T7 and T8instead of an interaction with conserved A6-T6 bases asobserved for the SAMecI

In our structure H2 helix (T26-T36) also binds to theDNA via T26 and N27 that interact with A6 Thisinteraction is well-conserved through different bacterialstrains E11 of the H1 helix and residue K3 of theN-terminal region may reinforce DNA recognition of thisextremity A parallel could be made with the A11 of theH1 residue of SAMecI which also contacts DNA

DISCUSSION

Monomer pathway has a significant role inthe repressorndashDNA binding

In the past it has been assumed that in b-lactam resistanceregulation system B licheniformis 749I BlaI-WT inter-acts as a preformed dimer with its operator (15) Thebinding constants of the full-length SAMecI and BLBlaIrepressors in interaction with palindromic operatingsequences have been measured in the range of tens ofnanomolar (9) Using chemical-shift mapping we havedetermined affinity constants for the BLBlaI and SAMecINTD with the different half DNA operators For all thecomplexes [SAMecI-NTD][12OPmecA] [SAMecI-NTD][12OP1blaP] [BLBlaI-NTD][12OP1blaP] and[BLBlaI-NTD][12OPmecA] binding constants are in thesame range of hundreds of micromolar that is 100 timeshigher than for the dimeric repressors The relatively lowdimerization constant of 25 mM (9) observed for BLBlaIhas suggested that both monomer and dimer pathwayswere possible Now taking into account our quantitativevalues of monomer-DNA dissociation constants and invivo concentrations of BLBlaI estimated to 2 mM (9) wecan conclude that the monomer pathway contributessignificantly to the BlaI repressor binding mechanismsin vivo

In various Winged HTH dimeric repressor systems asimilar monomerndashdimer equilibrium has been established(928) Two repression systems comparable to BlaI areknown LexA and Rep The DNA binding constant ofLexA-NTD (29) and Rep-NTD (30) are respectively about1 and 20 mM close to values determined for SAMecI-NTD and BLBlaI-NTD Moreover initial assays measur-ing the dimerization constant of the LexA repressor(2931) reported a Kd of about 10ndash50 mM The associationconstant of the full-length Rep repressor has beenevaluated to 3ndash5 nM as well When Rep monomers bindspecific DNA sequences the CTD is moved away andstable repressor-operator contacts can be establishedIf the specificity is not high enough the tail competeswith DNA and prevents DNAprotein non-specificassociationsThe monomer pathway could provide a method for

rapid localization of the binding site involving sliding ofthe protein along the DNA Theoretical considerationsand experimental evidence of protein sliding along non-specific DNA sequences are now well documented(32ndash35) This kind of search could combine 1D slidingwith 3D diffusion (36) driven by energetic differences forspecific and non-specific DNA-binding In our system thelocalization speed of the correct DNA-binding sitedepends on the competition of the monomer pathwaywith the dimer pathway Although strong cooperativeeffects in monomer association with DNA do not allow usto experimentally evaluate this process we note that sucha mechanism could play a role in this system Monomerswould be able to reach the adapted DNA motif using suchdiffusive processes and this pre-recognition step wouldallow a simpler contact between the two CTD Thisintermediate step might be useful for the protein toestablish correct and strong contacts

Corepression and adaptability

The presence of low affinity constants for the DNAbinding of the monomeric form can be correlated with theadaptability of the repressors for different operators Forexample the dimeric LexA repressor tightly binds thedifferent recA uvrB dinC and dinB DNA sequences Inthe case of BlaI the co-repression mechanism of homo-logous repressors in S aureus has been demonstrated (37)The Kd values measured in this work demonstrate that theBLBlaI-NTD protein is able to bind to both cognate andcrossed semi-operators in the same affinity range Thus itsuggests that low affinity complex formation might be anessential intermediate relay in binding variable regulationsequencesContrary to BLBlaI-NTD with mecA half-operator we

have shown that the affinity for blaP semi-operatingsequence is 10 times reduced for the SAMecI-NTDrepressor This parameter might corroborate observationsperformed on clinical S aureus isolates (3839) Indeed ithas been shown that the majority of oxacillin-resistantstrains contain deleted or mutated mecI genes while blaIsequences remain intact Thus in S aureus clinicalisolates the exclusive expression of MecI repressorappears to be a drawback in stress condition Indeed in

Nucleic Acids Research 2007 Vol 35 No 13 4391

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

this case bacteria are unable to respond rapidly TheBlaIBlaR1 system associated with the MecIMecR1machinery may enhance antibiotic resistance by improv-ing capability to respond Affinity of the monomer couldreflect the capacity to form some intermediate statesrequired during the induction process and the BlaI-NTDmight play a crucial role thanks to its inter-operatoradaptability

The dimerization of the repressor on the DNA involves atertiary rearrangement of the DNAndashmonomer complex

In the monomer pathway mechanism a first monomerbinds to the DNA exhibiting the monomer-DNAconformation described in this work This structure isindeed stabilized by a large number of proteinndashDNAinteractions that are systematically present in the dimerndashDNA interface (Table 3) Indeed the comparison betweenthe position of the monomeric and dimeric repressor onthe DNA gives a 308 rotation In this conformationalchange the overall position of the protein on the DNAremains the same that is the helix H3 stays in the maingroove but it shifts from one base (Figure 5) In thismodel the energetically favorable monomerndashDNA inter-face is destabilized upon the binding of another monomermolecule onto the DNAndashmonomer complex However itis easy to interpret the increase of the global affinity byconsidering the gain in enthalpy and entropy due to thepresence of two DNAndashmonomer interfaces and anadditional dimerization interface We note that someenergy is stored in the two monomerndashDNA interfaceswhen a dimerndashDNA complex is formed that can bepotentially released if the dimerization domain isdestabilized

Amonomerization of the repressors can explain thederepression mechanism

The values of the monomerndashDNA affinities measured inthis work (hundreds of micromolar) and the concentra-tions of repressors measured in the cell (2mM) (9) showthat monomers are not able to repress the genes In thepresence of b-lactam antibiotics the induction processmodifies the affinity of the protein repressor for its cognateDNA sequence The lower affinity measured for themonomer to its DNA-binding domain could then besufficient to explain the derepression of the geneHowever intermediate steps driving the release mecha-nism are not clearly understood

Biochemical investigations performed on the blaR1blaIblaZ b-lactamase regulation system (15) proposedthat proteolysis drives the signal transduction Indeed inStaphylococci the BlaR1 penicillin receptor is a membraneprotein containing a zinc metalloprotease motif Acylationof the sensor-transducer via penicillin binding triggers thesignaling mechanism leading to bla genes transcriptionThis event has been proposed to be the result of thecleavage of the dimeric S aureus BlaI repressor betweenresidues N101 and F102 Two induction models have beenproposed for the bla operator transcription underb-lactams stress conditions (1214) The first takes intoaccount the fact that a fourth gene as blaR2 in bla strainsencodes a key protein necessary for the CTD accessibilityimprovement (5613) The second considers that aninductor produced in the cytoplasm in stress conditionscan modulate the quaternary blocked conformation of theregulations elements In this way the inductor would beable to act as a proteolysis enhancer like in the TetRsystem (40) To sum up the lower affinity of the repressorduring the induction process could be the result of either a

Figure 5 Superposition of the S aureus MecI and the monomeric unit of the B licheniformis BlaI repressor (A) Side view of the dimeric S aureusMecI DNA binding domain (in green and blue) in complex with the bla sequence (PDB ID in orange) The monomeric BLBlaI obtained by de novodocking (in red) has been superimposed to MecI using only the conserved DNA recognition motif The B licheniformis BlaI DNA binding domains(in red) and the arrows point out the 308 rotation which is necessary to superpose the two monomeric BlaI-NTD and the two MecI-NTD domains(B) Side view of the S aureus MecI repressor (ribbon representation) in interaction with the 25 base-pair bla operating sequence The monomericunits are shown in different colors (light blue and green) (C) Orientation of the dimeric S aureus MecI repressor where each SAMecI-NTD unit hasbeen superimposed on the [BLBlaI-NTD][12OP1blaP] complex The 308 rotation described in panel A to superimpose the SAMecI and themonomeric BLBlaI-NTD creates a complete disruption of the dimerization domains in the case of a complete rigid molecule This twist can beconsidered as a starting point for more complex structural modification following the induction process

4392 Nucleic Acids Research 2007 Vol 35 No 13

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

proteolysis of the repressor leading to monomerizationor a structural change of the dimer that would place thetwo monomers in an unfavorable conformation leadingto a dimerndashDNA dissociation constant almost equal to amonomerndashDNA dissociation constant

The energetically most favorable monomerndashDNAconformation could represent a possible intermediate stateinduced in the inactivation process

For both schemes the structural study of the monomericBLBlaI-NTD bound to its DNA gives some informationon a possible intermediate state induced in the inactivationprocess and sheds light on the structural role of the C-terminal domain in the DNA-binding affinity

In both crystal structures of the SAMecI and SABlaI ininteraction with the bla and mec DNA sequence (56) thecleavage site is not easily accessible suggesting thatconformational changes may be necessary Furthermoreit has been observed that two forms open and closedcould co-exist allowing structural adaptation betweenmec and bla operators with different inter-N-terminaldomain distances (1041)

This structural flexibility may be permitted by the highdynamics observed for the dimerization part of therepressors (7) Here considering the structural differencesbetween the monomeric and dimeric interactions of BlaIwith DNA we propose a model where the dimer placed onDNA plays the role of a tense spring that can be releasedby a modification of the dimerization domains of therepressors This tense spring is composed of a transla-tional component similar to the one observed between theopen and closed form (6) and a supplementary rotationalcontribution of 308 During this release the monomerrepressor would slide 308 inside the major groove of theDNA until it reaches its equilibrium conformation (asdescribed in this article) Figure 5C displays the relativeposition obtained for the two CTDs in case of a completerigid structure and illustrates an incompatibility betweenthe twist of the two NTDs and conservation of an intactdimerization domain This twist displacement could inpractice modify the accessibility of the cleavage site byinvolving a modification of the secondary structurearrangement in the core of CTD and could result inbreakage of covalent bonds in this region This structuralmodification would be responsible for an irreversibleseparation of the two helices of the dimerization domainsubstantiating the possible existence of an inactivation-state DNA-binding domain conformation In the MexRrepressor study (42) two conformations have beenobserved as well On one hand the open state exhibitsoptimal DNA-binding domains interspaces which privi-leges the association with the operator On the other handa modification in between the DNA-binding domainsavoids repressor stowage on DNA Furthermore thissliding mechanism might be an explanation for a potentialrepressor ability to adjust contact between the closed andopen form

Recently published results (43) underline the fact thatmobility in the tertiary arrangement might be a usefulrequirement for thin regulation processes even in

eukaryotic systems Indeed the homeodomain of thetranscription factor Pdx1 is able to bind a 15-bp DNApromoter (with a consensus binding site) and adopt twoslightly different conformations in the same asymmetricunit Both complexes differ by a 248 rotation Additionalmodifications such as DNA curvature and N-terminallocal adjustments have been also raised by the authorSimilarly major structural rearrangements controlled

by small molecules have been highlighted for othersystems as hormone receptors (44) For the FadRtranscriptional regulator (45) interaction with DNA isdisrupted by the acyl-CoA binding Dramatic structuralrearrangements happen which leads the DNA recognitionhelices being separated by 72 A Another repressorinactivation process has been published concerningbacterial resistance to antibiotic Indeed the TetRrepressor is released in presence of tetracycline Thismolecule associated with Mg2+ plays the role of theinducer Inducer binding generates structural changes inthe CTD Thus a pendulum-like motion increasesthe separation of the attached DNA-binding domains(40) by 3 AThese models give strong support to the hypothesis that

an inducer or a blaR2 product participates to the initiationof C-terminal remodeling and propagation to theN-terminal parts (46) In presence of the effector specificand optimized contacts between the N-terminal part andthe DNA could be re-established by increasing theinterspacing between the TACATGTA motifs Thedifferences in the length of the interspaces between therecognition motifs could limit the DNA-binding domainsadaptation quality of SAMecI-WT to bla operatorTo conclude the weak affinities obtained for the NTD

and the monomerndashDNA structure demonstrate the crucialrole played by the CTD in an optimal positioning of theN-terminal units When a dimerndashDNA complex is formedsliding of the repressor on the DNA creates a tense springwhich can be released by the structural modificationsproduced by an inductor during the induction processThe monomerndashDNA structure described here may then beseen as an intermediate state during the monomer path-way repression scheme and also during induction afterdestabilization of the dimerization domain

ACCESSION NUMBERS

Coordinates for the model of the complex have beendeposited in the protein Data Bank (PDB ID code 2P7C)

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Online

ACKNOWLEDGEMENTS

The authors thank Catherine Bougault and Pierre Gansfor stimulating discussions and Dr Kevin Gardner at theUT Southwestern Medical Center in Dallas for providingscripts on the NMRView software This work wassupported by the lsquoFonds National de la Recherche

Nucleic Acids Research 2007 Vol 35 No 13 4393

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

Scientifiquersquo (FNRS) the lsquoCommunaute Francaise deBelgiquersquo (projet Tournesol) and the European Union FP6Integrated Project EUR-INTAFAR (Project LSHM-CT-2004-512138) under the thematic priority Life SciencesGenomics and Biotechnology for Health Funding to paythe Open Access publication charges for this article wasprovided by EUR-INTAFAR

Conflict of interest statement None declared

REFERENCES

1 SalernoA and LampenJ (1986) Transcriptional analysis of beta-lactamase regulation in Bacillus licheniformis J Bacteriol 166769ndash778

2 FudaCC FisherJF and MobasheryS (2005) Beta-lactamresistance in Staphylococcus aureus the adaptive resistance of aplastic genome Cell Mol Life Sci 62 2617ndash2633

3 WittmanV LinH and WongH (1993) Functional domains of thepenicillinase repressor of Bacillus licheniformis J Bacteriol 1757383ndash7390

4 Garcıa-CastellanosR MarreroA Mallorquı-FernandezGPotempaJ CollM and Gomis-RuthF (2003) Three-dimensionalstructure of MecI Molecular basis for transcriptional regulation ofstaphylococcal methicillin resistance J Biol Chem 27839897ndash39905

5 Garcıa-CastellanosR Mallorquı-FernandezG MarreroAPotempaJ CollM and Gomis-RuthF (2004) On the transcrip-tional regulation of methicillin resistance MecI repressor incomplex with its operator J Biol Chem 279 17888ndash17896

6 SafoM ZhaoQ KoT MusayevF RobinsonH ScarsdaleNWangA and ArcherG (2005) Crystal structures of the BlaIrepressor from Staphylococcus aureus and its complex with DNAinsights into transcriptional regulation of the bla and mec operonsJ Bacteriol 187 1833ndash1844

7 Van MelckebekeH VreulsC GansP FileeP LlabresGJorisB and SimorreJ-P (2003) Solution structural study of BlaIimplications for the repression of genes involved in beta-lactamantibiotic resistance J Mol Biol 333 711ndash720

8 GregoryPD LewisRA CurnockSP and DykeKG (1997)Studies of the repressor (BlaI) of beta-lactamase synthesis inStaphylococcus aureus Mol Microbiol 24 1025ndash1037

9 FileeP VreulsC HermanR ThammI AertsT De DeynPFrereJ and JorisB (2003) Dimerization and DNA bindingproperties of the Bacillus licheniformis 749I BlaI repressorJ Biol Chem 278 16482ndash16487

10 SharmaV HackbarthC DickinsonT and ArcherG (1998)Interaction of native and mutant MecI repressors with sequencesthat regulate mecA the gene encoding penicillin binding protein 2ain methicillin-resistant staphylococci J Bacteriol 180 2160ndash2166

11 FileeP DelmarcelleM ThammI and JorisB (2001) Use of anALFexpress DNA sequencer to analyze protein-nucleic acidinteractions by band shift assay Biotechniques 30 1044ndash1048

12 ZhangH HackbarthC ChanskyK and ChambersH (2001) Aproteolytic transmembrane signaling pathway and resistance tobeta-lactams in staphylococci Science 291 1962ndash1965

13 CohenS and SweeneyHM (1968) Constitutive penicillinaseformation in Staphyloccocus aureus owing to a mutation unlinkedto the penicillinase plasmid J Bacteriol 95 1368ndash1374

14 FileeP BenlafyaK DelmarcelleM MoutzourelisG FrereJBransA and JorisB (2002) The fate of the BlaI repressor duringthe induction of the Bacillus licheniformis BlaP beta-lactamaseMol Microbiol 44 685ndash694

15 WittmanV and WongH (1988) Regulation of the penicillinasegenes of Bacillus licheniformis interaction of the pen repressor withits operators J Bacteriol 170 3206ndash3212

16 ClarkeS and DykeK (2001) Studies of the operator region of theStaphylococcus aureus beta-lactamase operon J AntimicrobChemother 47 377ndash389

17 JohnsonB (2004) Using NMRView to visualize and analyze theNMR spectra of macromolecules Methods Mol Biol 278313ndash352

18 HilbersCW and WijmengaSS (1996) Nucleic acids spectrastructures and dynamics Encyclopedia of Nuclear MagneticResonance John Wiley amp Sons Chichester UK pp3346ndash3359

19 BonvinA (2006) Flexible protein-protein docking Curr OpinStruct Biol 16 194ndash200

20 NilgesM (1995) Calculation of protein structures with ambiguousdistance restraints Automated assignment of ambiguous NOEcrosspeaks and disulphide connectivities J Mol Biol 245645ndash660

21 DominguezC BoelensR and BonvinA (2003) HADDOCK aprotein-protein docking approach based on biochemical or bio-physical information J Am Chem Soc 125 1731ndash1737

22 PearlmanD and KollmanP (1991) Are time-averaged restraintsnecessary for nuclear magnetic resonance refinement A modelstudy for DNA J Mol Biol 220 457ndash479

23 BlackledgeM MedvedevaS PoncinM GuerlesquinFBruschiM and MarionD (1995) Structure and dynamics offerrocytochrome c553 from Desulfovibrio vulgaris studied by NMRspectroscopy and restrained molecular dynamics J Mol Biol 245661ndash681

24 CordierF CaffreyM BrutscherB CusanovichM MarionDand BlackledgeM (1998) Solution structure rotational diffusionanisotropy and local backbone dynamics of Rhodobacter capsulatuscytochrome c2 J Mol Biol 281 341ndash361

25 MortonC PughD BrownE KahmannJ RenzoniD andCampbellI (1996) Solution structure and peptide binding of theSH3 domain from human Fyn Structure 4 705ndash714

26 van DijkM van DijkA HsuV BoelensR and BonvinA (2006)Information-driven protein-DNA docking using HADDOCK it is amatter of flexibility Nucleic Acids Res 34 3317ndash3325

27 WallaceA LaskowskiR and ThorntonJ (1995) LIGPLOT aprogram to generate schematic diagrams of protein-ligand interac-tions Protein Eng 8 127ndash134

28 Mohana-BorgesR PachecoA SousaF FoguelD AlmeidaDand SilvaJ (2000) LexA repressor forms stable dimers in solutionThe role of specific dna in tightening protein-protein interactionsJ Biol Chem 275 4708ndash4712

29 KimB and LittleJ (1992) Dimerization of a specific DNA-bindingprotein on the DNA Science 255 203ndash206

30 IlangovanU WojciakJ ConnollyK and ClubbR (1999) NMRstructure and functional studies of the Mu repressor DNA-bindingdomain Biochemistry 38 8367ndash8376

31 SchnarrM PouyetJ Granger-SchnarrM and DauneM (1985)Large-scale purification oligomerization equilibria and specificinteraction of the LexA repressor of Escherichia coli Biochemistry24 2812ndash2818

32 GowersD WilsonG and HalfordS (2005) Measurement ofthe contributions of 1D and 3D pathways to the translocation ofa protein along DNA Proc Natl Acad Sci USA 10215883ndash15888

33 IwaharaJ and CloreG (2006) Detecting transient intermediates inmacromolecular binding by paramagnetic NMR Nature 4401227ndash1230

34 WinterR BergO and von HippelP (1981) Diffusion-drivenmechanisms of protein translocation on nucleic acids 3 TheEscherichia coli lac repressorndashoperator interaction kinetic mea-surements and conclusions Biochemistry 20 6961ndash6977

35 WinterR and von HippelP (1981) Diffusion-driven mechanisms ofprotein translocation on nucleic acids 2 The Escherichia colirepressorndashoperator interaction equilibrium measurementsBiochemistry 20 6948ndash6960

36 SlutskyM and MirnyL (2004) Kinetics of protein-DNAinteraction facilitated target location in sequence-dependentpotential Biophys J 87 4021ndash4035

37 McKinneyT SharmaV CraigW and ArcherG (2001)Transcription of the gene mediating methicillin resistance inStaphylococcus aureus (mecA) is corepressed but not coinduced bycognate mecA and beta-lactamase regulators J Bacteriol 1836862ndash6868

38 RosatoA KreiswirthB CraigW EisnerW ClimoM andArcherG (2003) mecA-blaZ corepressors in clinicalStaphylococcus aureus isolates Antimicrob Agents Chemother 471460ndash1463

4394 Nucleic Acids Research 2007 Vol 35 No 13

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from

39 RosatoA CraigW and ArcherG (2003) Quantitation of mecAtranscription in oxacillin-resistant Staphylococcus aureus clinicalisolates J Bacteriol 185 3446ndash3452

40 OrthP SchnappingerD HillenW SaengerW and HinrichsW(2000) Structural basis of gene regulation by the tetracyclineinducible Tet repressor-operator system Nat Struct Biol 7215ndash219

41 ClarkeS and DykeK (2001) The signal transducer (BlaRI) and therepressor (BlaI) of the Staphylococcus aureus beta-lactamaseoperon are inducible Microbiology 147 803ndash810

42 LimD PooleK and StrynadkaN (2002) Crystal structureof the MexR repressor of the mexRAB-oprM multidrug effluxoperon of Pseudomonas aeruginosa J Biol Chem 27729253ndash29259

43 LongoA GuangaGP and RoseRB (2007) Structural basis forinduced fit mechanisms in DNA recognition by the Pdx1 home-odomain Biochemistry 46 2948ndash2957

44 HeX ChowD MartickM GarciaK et al (2001) Allostericactivation of a spring-loaded natriuretic peptide receptor dimer byhormone Science 293 1657ndash1662

45 van AaltenD DiRussoC and KnudsenJ (2001) The structuralbasis of acyl coenzyme A-dependent regulation of the transcriptionfactor FadR EMBO J 20 2041ndash2050

46 VreulsC FileeP Van MelckebekeH AertsT De DeynPLlabresG MatagneA SimorreJ FrereJ et al (2004)Guanidinium chloride denaturation of the dimeric Bacillus licheni-formis BlaI repressor highlights an independent domain unfoldingpathway Biochem J 384 179

Nucleic Acids Research 2007 Vol 35 No 13 4395

by guest on May 3 2016

httpnaroxfordjournalsorgD

ownloaded from