Deciphering the Distinct Role for the Metal Coordination Motif in the Catalytic Activity of...

doi:10.1016/j.jmb.2009.08.064 J. Mol. Biol. (2009) 393, 788–802

Available online at www.sciencedirect.com

Deciphering the Distinct Role for the Metal CoordinationMotif in the Catalytic Activity of Mycobacteriumsmegmatis Topoisomerase I

Anuradha Gopal Bhat1, Majety Naga Leelaram1,Shivanand Manjunath Hegde1 and Valakunja Nagaraja1,2⁎
1Department of Microbiologyand Cell Biology,Indian Institute of Science,CV Raman Avenue,Bangalore 560 012, India2Jawaharlal Nehru Centre forAdvanced Scientific Research,Bangalore 560 064, India
Received 13 July 2009;received in revised form27 August 2009;accepted 27 August 2009Available online3 September 2009

*Corresponding author. E-mail [email protected] used: MstopoI, My

topoisomerase I; toprim, topoisomerEscherichia coli topoisomerase I; NTDdomain; CTD, carboxy-terminal domethylenediaminetetraacetic acid; STStopoisomerase site; EMSA, electrophassay; AAS, atomic absorption spec

0022-2836/$ - see front matter © 2009 E

Mycobacterium smegmatis topoisomerase I (MstopoI) is distinct from typicaltype IA topoisomerases. The enzyme binds to both single- and double-stranded DNA with high affinity, making specific contacts. The enzymecomprises conserved regions similar to type IA topoisomerases fromEscherichia coli and other eubacteria but lacks the typically found zinc fingersin the carboxy-terminal domain. The enzyme can perform DNA cleavage inthe absence of Mg2+, but religation needs exogenously added Mg2+. Onemolecule of Mg2+ tightly bound to the enzyme has no role in DNA cleavagebut is needed only for the religation reaction. The toprim (topoisomerase–primase) domain inMstopoI comprising theMg2+ binding pocket, conservedin both type IA and type II topoisomerases, was subjected to mutagenesis tounderstand the role of Mg2+ in different steps of the reaction. The residuesD108, D110, and E112 of the enzyme, which form the acidic triad in theDXDXE motif, were changed to alanines. D108A mutation resulted in anenzyme that is Mg2+ dependent for DNA cleavage unlike MstopoI andexhibited enhanced DNA cleavage property and reduced religation activity.The mutant was toxic for cell growth, most likely due to the imbalance incleavage–religation equilibrium. In contrast, the E112A mutant behaved likewild-type enzyme, cleaving DNA in a Mg2+-independent fashion, albeit to areduced extent. Intra- and intermolecular religation assays indicated specificroles for D108 and E112 residues during the reaction. Together, these resultsindicate that the D108 residue has amajor role during cleavage and religation,while E112 is important for enhancing the efficiency of cleavage. Thus,although architecturally and mechanistically similar to topoisomerase I fromE. coli, the metal coordination pattern of the mycobacterial enzyme is distinct,opening up avenues to exploit the enzyme to develop inhibitors.

© 2009 Elsevier Ltd. All rights reserved.

Keywords: topoisomerase I; mycobacteria; DNA relaxation; toprim domain;metal coordination
Edited by J. Karn
ress:

cobacterium smegmatisase–primase; EctopoI,, amino-terminalain; EDTA,, strongoretic mobility shifttrometry.

lsevier Ltd. All rights reserve

Introduction

DNA topoisomerases are an important class ofenzymes altering the topology of DNA by concertedbreakage and rejoining of the phosphodiester back-bone. The enzymes are vital in all the DNAtransaction processes such as replication, transcrip-tion, chromosome segregation, repair, andrecombination.1–4 Topoisomerases are classifiedinto two groups, type I and type II, depending ontheir ability to introduce transient single- or double-stranded breaks, respectively. Type I enzymes arefurther classified into types IA, IB, and IC that are

d.

http://dx.doi.org/10.1016/j.jmb.2009.08.064

789Toprim Mutants of Mycobacterial Topoisomerase I

structurally and mechanistically dissimilar and havedifferent set of domains.2,5 Type IA topoisomerasesform 5′ phosphotyrosine covalent adducts with thecleaved phosphodiester backbone releasing a free 3′hydroxyl group while the enzymes from other twosubgroups form 3′ phosphotyrosine covalentadducts during the reaction. Bacterial topoisome-rases I and III belong to the type IA group while poxvirus6 and human topoisomerase I7 belong to thetype IB group. The DNA relaxation reaction by thetype IA topoisomerases is Mg2+ dependent, and theconformational changes required for relaxationreaction are suggested to be essential for the reactionsteps of strand passage, religation, and productrelease.8

The toprim (topoisomerase–primase) domain iswidely distributed among bacterial, archeal, andeukaryotic proteins, functioning as the metal coordi-nation site in enzymes that catalyze the phosphodie-ster bond cleavage/formation.9–14 As the nameimplies, toprim domain is found in bacterial typeIA/type II DNA topoisomerases and DnaG-likeprimases. In addition, it is also found in OLD familynucleases, RecR family ofDNArepair proteins,9 DNApolymerase I,15–17 and RNase M5.18 The toprimdomain is found in combinationwith other functionalmotifs exhibiting different substrate binding require-ments and functions for the enzymes.9,14 A conservedset of acidic amino acids within the domain partici-pate in metal ion coordination.Mg2+ is the preferred divalent metal ion in most of

the enzymatic systems. It is an essential co-factor thatcan have either structural or catalytic functions in thecells. Based on the studies with Escherichia coli DNAtopoisomerase I (EctopoI), Mg2+ was found to beessential for the relaxation activity.19 Apart from itsrole in religation, where the 3′ OH group of thecleaved DNA is possibly activated by the metal ion toreattack the phosphotyrosine,20 the metal ion isimportant for conformational changes in the enzymeduring the reaction. A conserved DXDXE/D motifwithin the toprim domain is involved in Mg2+

coordination. The acidic triad D111XD113XE115 motifmutants to alanines have been well characterized inEctopoI. These site-directed mutants caused onlypartial loss of relaxation activity of the enzyme.However, the double and triple mutants of the acidictriad to alanines exhibited more than 90% loss ofactivity, highlighting their role in catalysis throughMg2+ coordination.13,21DNA topoisomerase I from Mycobacterium smegma-

tis (MstopoI) exhibits distinct features from other typeIA topoisomerases.22 The enzyme can recognize bothsingle- and double-stranded DNA with high affinityand makes sequence-specific contacts with DNAduring its reaction cycle.23–25 Taking advantage of itssite-specific binding, extensive footprinting studieshave been carried out.25 Based on these studies, theenzyme is proposed to employ an enzyme-bridgedmechanism to remove negative supercoils fromDNA,similar to the reaction mechanism originally putforward for the 67-kDa N-terminal fragment ofEctopoI.8 The sequence specificity for MstopoI is

imparted by its amino-terminal domain (NTD),which carries out the sequence-specific DNA binding,cleavage, and religation reactions, characteristic of theenzyme, but lacks relaxation activity on its own. Thecarboxy-terminal domain (CTD) is also indispensablefor the relaxation activity of the enzyme. The CTDbinds to DNA in a sequence-independentmanner andappears to play an important role during strandpassage step of the reaction.26 Multiple sequencealignment of type IA topoisomerases revealed severalconserved domains and residues in the NTD ofmycobacterial topoisomerase I. In contrast, CTD ofthe enzyme is highly varied and is quite distinct fromother well-characterized type IA enzymes. The char-acteristic three zinc fingers found at the CTD inEctopoI and other eubacterial type IA enzymes areabsent in mycobacterial topoisomerase I.27

To understand the role of Mg2+ coordination in thefunction of the enzyme, we have examined the acidictriad inMstopoI. Site-directedmutagenesis of the firstacidic residue in the toprim domain of the enzymeshowed lethal phenotype. A similar mutation inEctopoI was not toxic and retained activity closest towild-type enzyme,10,13 suggesting subtle differencesin the pattern of Mg2+ coordination between the twoenzymes. As a result, we have carried out detailedanalysis of the acidic triad mutants in MstopoI tounderstand their role in different steps of DNArelaxation.

Results

Second transesterification reaction bytopoisomerase I is Mg2+ dependent

DNA relaxation by MstopoI is Mg2+ dependent(Fig. 1a, lanes 6–9). In the absence of Mg2+, thenegatively supercoiled DNA substrate was trans-formed to nicked DNA (Fig. 1a, lanes 2–5) withincreasing enzyme units. The appearance of nickedDNA indicates the event of first transesterificationreaction, revealing thatMstopoI is competent in DNAcleavage without Mg2+. This is confirmed by oligo-nucleotide cleavage in the absence of Mg2+ (Fig. 1b,lane 2; Fig. 1c, lane 3). Two enzyme preparations, onedialyzed extensively with ethylenediaminetetraaceticacid (EDTA) and the other denatured and refolded inabsence of Mg2+, also retained DNA cleavage activity(see later section, Fig. 9). However, the secondtransesterification reaction, which leads to ligationof the DNA strands, is Mg2+ dependent. To addressthe requirement of Mg2+ in intramolecular religation,MstopoI was allowed to form a covalent complexwith labeled DNA [5′-end-labeled 32-mer oligonucle-otide containing strong topoisomerase site (STS)annealed to 32-mer complementary strand]. WhenMg2+ was added to facilitate the religation, thecleaved product diminished with increasing Mg2+

(Fig. 1b, lanes 3–8). To assess intermolecular religation(Fig. 1c),we carried outDNAcleavage assayswith thesame 32-mer oligonucleotide containing STS to

Fig. 1. Mg2+ requirement at different steps of DNA relaxation byMstopoI. (a) Formation of nicked DNA in the absenceof Mg2+ and DNA relaxation in the presence of Mg2+: Lane 1 contains supercoiled, nicked, and linear DNA controls withno enzyme. Linear DNA in the control lane is generated by restriction enzyme digestion and mixed with other DNAcontrols. Lanes 2–5 (no Mg2+ added) and 6–9 (Mg2+ supplemented) contain increasing enzyme concentrations (20–80 nMas shown). (b) DNA cleavage with increasingMg2+ : 32-mer oligonucleotide containing STSwas annealed to unlabeled 32-mer non-STS complementary strand. This substrate DNA (lane 1) was incubated with 200 nM enzymewithout Mg2+ (lane2) and with varying concentrations of Mg2+ from 0.1 to 10 mM (lanes 3–8). (c) Intermolecular religation: Experimentaldesign. To assess intermolecular religation, we carried out DNA cleavage assays with topoisomerase I using 32-mer STSoligonucleotide (c, lane 1), which would generate 19-mer-labeled product (c, lane 3) and 13-mer protein–DNA covalentintermediates. A partial duplex DNA of 24-mer, complementary to the 13-mer cleaved product and 5′-end-labeled 11-mer(c, lane 2), identical in sequence to the 32-mer cleavage product was added to the reaction mixture subsequent to cleavage.This generates a duplex DNA with a nick residing at the covalent adduct site, a substrate for intermolecular religation,resulting in appearance of labeled 24-mer as the religation product, which is Mg2+ dependent (c, compare lanes 4 and 5).The reaction is performed in the absence of Mg2+ in lane 4 while 5 mM Mg2+ was included in lane 5.

790 Toprim Mutants of Mycobacterial Topoisomerase I

generate 19-mer-labeled product (Fig. 1c, lane 3) and13-mer protein–DNA covalent intermediate. 5′-end-labeled 11-mer, preannealed to 24-mer complemen-tary strand,was used as acceptorDNAalongwith the13-mer protein–DNA covalent intermediate (Fig. 1c),generating duplex DNA with a nick at the covalentadduct site to facilitate intermolecular religation.

The generation of labeled 24-mer as a result ofintermolecular religation occurred only in thepresence of Mg2+ (Fig. 1c, compare lanes 4 and 5).MstopoI exhibited [Mg2+]-dependent religation witha temperature optimum of 37 °C (Fig. S1a and c)and was stable even up to 120 min during thereaction (Fig. S1e).


Substitution of D108 with alanine leads to in vivotoxicity

Multiple sequence alignment ofMstopoIwith othertype IA topoisomerases revealed conservation ofthree acidic residues DXDXE within the toprimdomain (Fig. 2a). Mutagenesis of the acidic residuesto alanine was carried out as described in Materialsand Methods to address the role of the individualamino acids in Mg2+ coordination and consequentenzyme activity. Surprisingly, a clone containing themutation in the first acidic amino acid (MstopoID108A)did not generate transformants in eitherE. coliBL21 or BL26 strains. This could be due to in vivotoxicity caused by the expression of the mutantenzyme (Fig. 2b, sector 2). To test this observationand also to ensure highly regulated expression of themutant protein, we introduced T7 lysozyme-expres-sing vectors to these strains harboring MstopoID108A. T7 lysozyme is a natural inhibitor of T7RNA polymerase.28 The lethal effect of MstopoID108A was rescued with the strain harboring eitherpLysS or pLysE plasmid (Fig. 2b, sectors 4 and 5,respectively). In contrast, MstopoI E112A did notshow any lethal effect. The second acidic residue in

the toprim domain was also mutated to alanine(MstopoID110A), but themutant protein could not becharacterized further due to its inconsistent levels ofexpression and associated insolubility problems.

Purification and characterization of the mutants

The mutant proteins were purified by two columnchromatographic steps following the procedure de-scribed for MstopoI.26 The purified proteins weresubjected to CD and fluorescence spectroscopy toevaluate alterations in their secondary or tertiarystructures. From the far-UVCD spectral data (Fig. S2)and intrinsic tryptophan fluorescence (Fig. S3), itappears that there are no major secondary or tertiarystructural changes in MstopoI D108A. The mutanthad similar spectral profile comparable to MstopoIirrespective of the presence or absence of Mg2+. Thisalso indicated that the in vivo toxicity is not aconsequence of any misfolding in MstopoI D108Adue to the mutation. These results also suggest animportant role for the D108 residue.The susceptibility of MstopoI and its mutant forms

to proteolysis was compared in the presence of Mg2+.The tryptic profiles of the enzymes revealed thatMstopoI D108A was more susceptible to cleavageeither in the presence or in the absence of Mg2+.MstopoI and MstopoI E112A showed similar trypticprofiles at different [Mg2+] (Fig. 3).

Fig. 2. (a) Domainal organization of MstopoI: Con-served motifs are shown as dark boxes. NTD comprisesthe metal coordinating acidic triad DXDXE in the toprimdomain and the CAP domain or the cleavage domaincomprising the active-site ITYMRTDS motif. The putativeactive-site tyrosine 339 is shown in asterisk. Multiplesequence alignment of type IA topoisomerases fromdifferent organisms was carried out to illustrate theconserved acidic residues DXDXE displayed in black-colored shaded boxes with asterisks on the topmost panel.The alignment was performed with ClustalW version 2.0software and analyzed using GeneDoc program. Repre-sentative members of type IA topoisomerases for themultiple sequence analysis included the following: type Itopoisomerases from Mycobacterium tuberculosis (Mt),Mycobacterium bovis (Mb), M. smegmatis (Ms), Shigellasonnei (Ss), E. coli (Ec), Salmonella typhi (St), Staphylococcusaureus (Sa), Bacillus subtilis (Bs), Clostridium acetobutylicum(Ca), Helicobacter pylori (Hp), and type III topoisomerasesfrom E. coli (Ec topo III) and S. cerevisiae (Sc topo III). (b)Growth patterns of E. coli BL21 (DE3) cells overexpressingMstopoI and MstopoI D108A. The mutant exhibited invivo toxic phenotype that was overcome with theregulation of T7 RNA polymerase in E. coli BL21 harboringeither pLysS or pLysE plasmid. The respective competentcells were individually transformed with the plasmidDNA, with pPVN123 encoding either wild-type enzymeor D108A. Plasmid pRSETA was also transformed asvector control. After revival, the cells were streaked on LBagar plate and incubated at 37 °C to show their respectivegrowth. Sector 1, MstopoI in E. coli BL21; sector 2, MstopoID108A in E. coli BL21; sector 3, MstopoI in E. coli BL21pLysS; sector 4, MstopoI D108A in E. coli BL21 pLysS;sector 5, MstopoI D108A in E. coli BL21 pLysE; sector 6,pRSETA vector control in E. coli BL21.

Fig. 3. Susceptibility of MstopoI and its mutants to proteolysis: Silver-stained gel with the tryptic profile of MstopoIand its mutant forms in the presence of 0, 5, 10, and 20 mMMg2+. Trypsin (5 ng) was used for 4 μg of each protein, and theproteolysis was carried out at 24 °C for 20 min and resolved on 8% SDS PAGE.


DNA relaxation assays using negatively super-coiled pUC18 DNA substrate revealed that MstopoID108A and MstopoI E112A mutations had a 20- and5-fold reduction in their specific activities, respective-ly, compared to MstopoI (Fig. 4 and Table 1).Electrophoretic mobility shift assays (EMSAs) werecarried out to assess the DNA binding characteristicsofMstopoImutants. Although theDNA relaxation bymetal coordination motif mutants was affected tovarious extents, EMSAs with 5′-end-labeled oligonu-cleotides containing STS revealed that the DNAbinding by these enzymes did not require Mg2+. Todetermine their affinities to the specific DNAsequence, we incubated increasing concentration ofproteins with the DNA, and the DNA–proteincomplexes were analyzed by EMSA. From the plotsshown in Fig. 5, the Kd values were determined for

Fig. 4. DNA relaxation activity with MstopoI and its mut500 ng of negatively supercoiled pUC18 DNA as substrate. Rethe respective enzymes from 10 to 1000 nM enzyme concentratsupercoiled DNA with no enzyme. One unit is defined as tsupercoiled pUC18 in 30 min at 37 °C.

both the mutants. The DNA binding affinities ofMstopoI D108A and MstopoI E112A were compara-ble to that of MstopoI, indicating that the mutationsdid not alter DNA binding properties of the enzyme(Table 1).

MstopoI D108A requires Mg2+ for DNA cleavage

DNA cleavage is Mg2+ independent in the case ofMstopoI23 with an optimumcleavage at 2.5 and 5mMMg2+. The DNA cleavage activity was reduced in thepresence of 10 and 20 mM Mg2+ (Fig. 6, lanes 2–6).DNA cleavage characteristics of MstopoI and itsmutant forms were compared at different timeintervals of 0, 5, 10, 20, and 30 min, with varyingconcentrations of Mg2+ (Fig. 6). No detectable DNAcleavage was observed in the case of MstopoI D108A

ant forms: Enzymes were serially diluted and assayed onlaxation reaction corresponds to varying concentrations ofions as mentioned. In each case, the first lane is negativelyhe amount of enzyme that relaxes 50% of the 500 ng of

Table 1. Specific activity and affinity of proteins

Protein Specific activity (U/mg) Kd (nM)

MstopoI 2.0×105 2.41D108A 0.1×105 7.87E112A 0.4×105 7.55

Fig. 5. DNA binding of MstopoI, MstopoI D108A, andMstopoI E112A proteins to specific single-stranded DNA:EMSAs were carried out by incubating 5′-end-labeled 32-mer containing STS with varying concentrations ofproteins from 1 to 200 nM. The graphs depict thepercentage of bound oligonucleotide with their respectiveEMSA gels in the inset, and their Kd values aresummarized in Table 1.


in the absence of Mg2+ but the mutant exhibitedthreefold stimulation at higher concentrations of themetal ion compared toMstopoI (Fig. 6, lanes 7–11). Incontrast, MstopoI E112A retained DNA cleavageactivity in the absence of Mg2+, although the extentof cleavage was reduced. However, cleavage wascomparable with MstopoI at different concentrationsof the metal ion (Fig. 6, lanes 12–16). From theseresults, it appears that MstopoI D108A is severelycompromised in DNA cleavage, while MstopoIE112A is only marginally affected.

Effect of the toprim domain mutations oninter- and intramolecular religation

Intermolecular religation assays were performedwith MstopoI and its metal coordination motifmutants at 5 and 10 mM [Mg2+]. In spite of excessMg2+ supplementation, MstopoI D108A revealed a30% and 40% reduction in religation activity at 5 and10mMMg2+, respectively, compared toMstopoI (Fig.7a and b). Intramolecular religation was drasticallyaffected in MstopoI D108A, as the 19-mer cleavedproduct increased with increasing Mg2+ concentra-tion (Fig. 8, lanes 5 and 6). Quantitations of thecleavage and religation products indicated that thecleavage–religation equilibrium was altered, imply-ing a role for D108 in religation. On the other hand,MstopoI E112A was equally efficient in intermolecu-lar religation and was affected to only about 10% inintramolecular religation when compared toMstopoI(Fig. 8, lanes 7 and 8).

Intrinsically bound Mg2+ is not required for DNAcleavage

The results presented above reinforce the require-ment of Mg2+ in the second transesterificationreaction. The absolute need of Mg2+ for the firsttransesterification by MstopoI D108A and Mg2+-independent DNA cleavage by MstopoI led us toanalyze intrinsically bound Mg2+ in the proteinpreparations.MstopoI and itsmutantswere subjectedto atomic absorption spectrometry (AAS) studies,which revealed that themole ratio ofMstopoI toMg2+

is 1:1. The enzymes were purified in a buffer contain-ing 10 mM EDTA to chelate out Mg2+, if any, andagain subjected to AAS. This study revealed thepresence of tightly bound Mg2+ in MstopoI prepara-tions with a mole ratio near 1. Unexpectedly, inMstopoI E112A, there was no bound Mg2+ (Table 2).Surprisingly, MstopoI D108A, which showed Mg2+

dependency for DNA cleavage unlike MstopoI orMstopoI E112A, also had 1mol ofMg2+ tightly boundto 1 mol of the protein. To address the role ofintrinsically bound Mg2+ in DNA cleavage, we

simultaneously prepared MstopoI and the mutantproteins by denaturation and subsequent renatur-ation to obtain proteins devoid of bound Mg2+.Atomic absorption studies of the enzymes revealedthe absence of bound Mg2+ in all the renaturedsamples. TheseMstopoI andMstopoI E112Aproteins,devoid of bound Mg2+, were able to carry out DNAcleavage efficiently (Fig. 9), confirming that Mg2+

coordination is not a prerequisite for DNA cleavage.MstopoI D108Awas compromised in DNA cleavage,irrespective of the presence or absence of tightlybound Mg2+. In spite of tightly bound Mg2+ in theMstopoI preparations, it did not rely on the boundMg2+ for DNA cleavage. The bound Mg2+ by itselfcould not support overall relaxation reaction orreligation (see Fig. 1a, lanes 2–5; Fig. 1b, lane 2; Fig.

Fig. 6. Cleavage of 5′-end-labeled single-stranded DNA byMstopoI and its mutant forms in the absence and presence ofMg2+: 5′-end-labeled 32-mer oligonucleotides containing STS were used as cleavage substrate to yield a 19-mer-labeledproduct. Cleavage at different time intervals wasmonitored by arresting an aliquot of the reactionmixture at 0, 5, 10, 20, and30min. Lane 1, control oligonucleotidewith no enzyme; lanes 2–6, cleavagewithMstopoI; lanes 7–11, cleavagewithMstopoID108A; lanes 8–16, cleavage with MstopoI E112A. The right panel depicts the quantitation of cleavage with time at differentconcentrations ofMg2+. To quantify DNA cleavage, we considered as 100% themaximum cleavage bywild-type enzyme for30 min at each Mg2+ concentration. The cleavage with mutants at the respective Mg2+ concentration was normalized to thewild type.


1c, lane 4). Nicked DNA intermediates are formed inrelaxation assays without Mg2+ supplementation(Fig. 1a, lanes 2–5). These results show the indispens-able role of the externally added second molecule ofMg2+ in overall reaction.

MstopoI model to reveal Mg2+ coordination

The NTD of MstopoI shows 56.6% similarity withEctopoI.29 A homology model of MstopoI was builtbased on the well-studied EctopoI structure tounderstand the Mg2+ coordination by the enzymes,which would account for the above observations.8

The overlay structure for the NTD of MstopoIincluded all the catalytic domains along with thetoprim domain (Fig. S4a–c). The two-metal-ionmodel

has been suggested for topoisomerases that harbor thetoprim domain. The twoMg2+ ions have to optimallycoordinate and are supposed to be appropriatelyplaced at a distance limit between 3 and 4Å from eachother to facilitate the phosphoryl transfer.20 Theoverlapping structures of MstopoI on EctopoI tem-plate, restricted to the respective acidic triads and theactive-site tyrosine, were obtained to understand theoccupancy of the metal ion in the active center (Fig.10a). In the MstopoI model, only a single Mg2+ ionpossessing an atomic radius of 0.72 Å could beaccommodatedwithin the acidic triad proximal to theactive-site tyrosine (Fig. 10b). The positioning of theacidic triad with respect to Mg2+ ion appears to besimilar to the structure of Saccharomyces cerevisiaetopoisomerase II enzyme bound to DNA.30 In

Fig. 7. Intermolecular religationwithMstopoI andmutants: The experimental design is as depicted in Fig. 1c. The 5′-end-labeled 32-mer single-strandedDNA is the substrate that yields 5′-end-labeled 19-mer and 13-mer covalent enzymeadduct asproducts. The 13-mer covalent adduct is the substrate for religation that is complementary to the pre-annealed 24-mer and 5′-end-labeled 11-mer partial duplex to generate labeled 24-mer as religation product. Religation was initiated with 5 mM (a)and 10 mM (b) Mg2+, respectively. In both figures: lane 1, control oligonucleotide (F); lane 2, cleavage control without Mg2+

(C) inwhich cleavage reaction is carried out for 15min prior to the religation reaction; lane 3, religation at 0min (R); lanes 4–7,religations at 15, 30, 45, and 60 min (R). Cleavage is carried out in the presence of 5 mM Mg2+ as MstopoI D108A iscompromised in cleavage in the absence of Mg2+. Right panel is the graphical quantitation of religation reactions ofMstopoI,MstopoI D108A, and MstopoI E112A proteins. The low-intensity bands seen above and below 19-mer are likely to becontaminants and not the products of ligation.


Fig. 8. Intramolecular religation with MstopoI and mutants: 5′-end-labeled 32-mer oligonucleotides containing STSwere incubated with 200 nM of each protein in the presence of 5 mMMg2+ to yield 19-mer cleaved product. After 15 minof cleavage reaction, the intramolecular religation was initiated by increasing the Mg2+ to a final concentration of 10 mM.Lanes 1 and 2 are free oligonucleotides incubated with 5 and 10 mM Mg2+; lanes 3, 5, and 7 are cleavage reactions withMstopoI, MstopoI D108A, andMstopoI E112A proteins, respectively. Similarly, lanes 4, 6, and 8 reveal the intramolecularreligations with MstopoI, MstopoI D108A, and MstopoI E112A, respectively. The amount of cleavage obtained in thepresence of 5 mM Mg2+ for 15 min is considered as 100% (lanes 3, 5, and 7). The percentage of decrease/increase incleavage product at 10 mM Mg2+ was normalized to the respective values obtained with 5 mM Mg2+.


addition, the interatomic distance from the metal ionto the coordinating residue was considered to have aminimal optimumdistance of over and above 2Å. Thepossible interatomic distances were analyzed in themutant forms of both EctopoI and MstopoI (Fig. 10c).From the modeling analysis, it appears that bothEctopoI D111A/MstopoI D108A and EctopoID113A/MstopoI D110A would hold the Mg2+ in atight surrounding formed by the other two negativelycharged carboxyl groups of the triad along with thehydroxyl group of the active-site tyrosine in aprotected space (Fig. 10c). In contrast, EctopoIE115A/MstopoI E112A has a weaker surroundingto protect the bound Mg2+ with the two neighboringacidic groups that remain far apart (Fig. 10c), probablyreleasing the metal ion to the microenvironment.

Discussion

A number of enzymes involved in DNA/RNAtransaction processes coordinate Mg2+ ion for catal-ysis. Topoisomerases belonging to both type IA andtype II class also require Mg2+ for catalysis and wereconsidered to follow two-metal-ion-mediated catalyt-ic mechanism.20,21,31–33 In the two-ion model forcatalysis by nucleases, polymerases, and a numberof other DNA/RNA transaction enzymes, the firstmetal ion is needed to activate the nucleophile and the

Table 2. Mg2+ AAS

Protein Sample dialyzed against buffer Mole ratio Cleavage

Wt 1 mM EDTA 1.1±0.2 +Wt 10 mM EDTA 0.77±0.2 +D108A 10 mM EDTA 0.83±0.1 −E112A 10 mM EDTA 0.00 +Wt 10 mM EDTA (refolded) 0.00 +D108A 10 mM EDTA (refolded) 0.00 −E112A 10 mM EDTA (refolded) 0.00 +

Results were obtained from two individual sets of experiments.

second one is necessary to stabilize the pentacovalenttransition state and the leaving group.18,34–36 In thetwo transesterification reactions catalyzed by topoi-somerases,Mg2+ can participate in nucleophilic attackby hydroxyl group of tyrosine and stabilize thetyrosinate and the transition state as well as thenegatively charged leaving group.20,31 During catal-ysis, the twoMg2+ ions are suggested to bemobile andpossibly undergo coordination rearrangements re-quired for proper substrate orientation that occursduring the cleavage and religation steps. Importantly,in addition to the catalytic steps, Mg2+ also seems totrigger conformational changes in these enzymesduring the reaction.20 In the case of DNA gyrase,Mg2+ is required for the activation of the nucleophile,stabilization of the pentacovalent transition state, andactivation of the 3′ OH group for religation.31 Incontrast, in type IA enzymes, Mg2+ does not seem tobe required for DNA cleavage but important forconformational changes in the enzyme and religationreaction.20,37

Fig. 9. DNA cleavage assays with AAS proteinsamples prepared by denaturation and renaturation withno bound Mg2+: 5′-end-labeled 32-mer oligonucleotidescontaining STS were used as cleavage substrate to yield a19-mer-labeled product. Lane 1, oligonucleotide with noenzyme; lanes 2 and 3, cleavage with 50 and 100 nMMstopoI; lanes 4 and 5, cleavage with 50 and 100 nMMstopoI D108A; lanes 6 and 7, cleavage with 50 and100 nM MstopoI E112A.

Fig. 10 (legend on next page)



In this study onmutational analysis of MstopoI, wehave delineated further the importance of Mg2+

coordination by comparing the results obtainedwith the detailed study carried out on EctopoI. Onenotable property where the enzymes differ is withrespect to themode ofMg2+ binding. EctopoI binds totwoMg2+ per molecule, but the enzyme preparationswere devoid of tightly boundMg2+.19,21,27 In contrastto this, theMstopoI contained one tightly boundMg2+

in the purified preparations. The tightly bound Mg2+

had no role in DNA cleavage as the enzyme strippedof the metal ion retained the efficient cleavageproperty. From these results, we conclude that Mg2+is not necessary for nucleophilic activation during thecleavage step in the case ofMstopoI. This is consistentwith the structure of the E. coli topoIII–DNA complexin cleavage-competent state, in which Mg2+ wasabsent, suggesting a cleavage mechanism that doesnot involve the metal ion.38 The second Mg2+

molecule required for religation and overall relaxa-tion reaction in MstopoI appears to be loosely boundand stabilized only in the presence of DNA, a featurecommon to many other enzymes.The elucidation of the three-dimensional structure

of the 67-kDa NTD of EctopoI8 revealed the presenceof the acidic triad, D111, D113, and E115, proximal tothe active-site tyrosine 319.19 The DXDXE/D motif isconserved in all the known type IA and type IItopoisomerases and is required for catalysis.39,40 Thetoprimdomain comprising the acidic amino acid triadis the Mg2+ coordinating motif involved in acid–basecatalysis.20 Earlier studies showed that Mg2+ is notrequired for DNA cleavage or for the formation ofcovalent protein–DNA intermediate but is essentialfor religation.37,41 Site-directed mutants of the indi-vidual amino acids D111A, D113A, and E115A of theacidic triad in EctopoI resulted only in partial loss ofenzyme activity. In two separate studies carried outby Wang and Tse-Dinh's groups, the mutation in thefirst residue of the triad (D111A) resulted in themutant enzyme with modest reduction in overallrelaxation activity.10,13 However, double and triplemutants did not exhibit any detectable enzymeactivity due to the loss of Mg2+ binding.21

Mutational analysis of the conserved acidic triad inMstopoI described in this study exhibits differentproperties compared to the EctopoI mutants dis-cussed above. MstopoI D108A had no detectableDNA cleavage activity in the absence of Mg2+ unlikethe wild-type enzyme. In intermolecular religationreaction, upon addition of Mg2+, the cleaved 19-merproduct increased in the case of the mutant butdecreased in wild type. The mutant was severelycompromised in intramolecular religation. Together,these results imply that theD108 residuemight have adual role both in holding or positioning of the 3′ OH

Fig. 10. (a) Overlapping structures of the acidic triads andviewed using PyMOL software. The active-site tyrosine and thD111/D113/E115 and Y339/D108/D110/E112, respectively. (motif: Metal coordination residues at the active center of Mstoaccommodation of a Mg2+ metal ion. (c) Interatomic distancesEctopoI and MstopoI: Alanine mutations in EctopoI E115A an

group of the cleaved strand and in its subsequentactivation for the second transesterification reaction.Due to its inability to hold on to the 3′ end of the nickby the removal of carboxylate, the enzyme iscompromised in the religation reaction accountingfor the observed in vivo toxicity. The mutants of E. coliand Yersinia pestis topoisomerases, EctopoI D111Nand YptopoI D117N (both are first acidic residues ofthe triad), were also found to be lethal,42 possibly dueto the very similar mechanism. The co-crystalstructure of E. coli topoisomerase III also revealedthe importance of the first acidic residue, D103(equivalent to D108 in MstopoI), in interacting withthe oligonucleotide in the cleavage-competent state.38

Thus, it appears that in all these enzymes, the firstacidic residue has a critical role in the overall reaction.Surprisingly, MstopoI D108A was dependent on

Mg2+ for DNA cleavage, not typically required for atype IA enzyme at this step. The mutant was moresusceptible to tryptic digestion and showed distinctproteolytic cleavage pattern. However, it did notreveal any significant alteration in the secondary ortertiary structure in the presence of Mg2+ in CDspectroscopy and fluorescence quenching experi-ments. The change in the fluorescence emissionintensity seen upon addition of DNA in MstopoID108A (compared to the wild-type enzyme) possiblyindicates altered Mg2+-dependent active conforma-tion. This alteration in fluorescence emission in thepresence of DNA is not because of compromisedDNA binding, as the Kd value for the DNA bindingfor the mutant was comparable to MstopoI. Therequirement for Mg2+ in DNA cleavage, therefore,suggests an altered Mg2+ coordination scheme and adifferent reaction mechanism that may involvechanges in the proton relay in the mutant during theacid–base catalysis.43 Together, these observationssupport the idea of movement of Mg2+ from a moreburied environment to an exposed site as a conse-quence of the structural and electronic changes thatoccur during DNA cleavage.20

From the present studies, it appears that the E112residue in MstopoI is involved in holding/position-ing the strand, both for cleavage and for religation,enhancing the overall efficiency of cleavage andreligation. The mutant MstopoI E112A exhibitedreduced DNA cleavage in the absence of Mg2+ buthad little effect on the efficiency in the presence ofMg2+. In addition, the mutant was efficient inintermolecular religation and exhibited a marginaldecrease in intramolecular religation. It seems thatthe E112 residue may have a minor role in holdingthe cleaved strand for the second transesterificationreaction and is not involved directly in activation ofthe nucleophile. Thus, two of the residues thatcoordinate Mg2+ may have distinct roles during

the active-site tyrosine residues of EctopoI and MstopoI,e acidic triad in EctopoI and MstopoI correspond to Y319/b) Accommodation of Mg2+ within the metal coordinationpoI depicting the interatomic distances in angstroms afterin the site-directed alanine mutants of the acidic triad ind MstopoI E112A indicate lack of tightly bound Mg2+.


target capture/strand holding. Co-crystal structuresof topoisomerase I–DNA in cleaved state wouldprovide better understanding of the role of individ-ual amino acids of the toprim domain in strandcapture and the subsequent steps of the reaction.In order to better understand our results and to

correlate with the mechanism of action of other typeIA enzymes, we have modeled the toprim and theactive-site residues based on the crystal structure ofthe 67-kDa NTD fragment of EctopoI8 (Fig. 10). Inparallel, the MstopoI–DNA complex was modeledbased on the EctopoIII–DNA complex38 (Fig. S5). Theactive-site tyrosine and the three acidic residues of thetoprim domain involved inMg2+ ion coordination areoptimally positioned in a near-identical fashion inboth EctopoI and MstopoI (Fig. 10). A remarkablysimilarMg2+ coordination scheme is seen in the case ofS. cerevisiae topoisomerse II.44 In all these cases, out ofthe two metal ions, only one could be positionedwithin thepocket at appropriate interatomic distances.This is the tightly bound ion observed in the case ofMstopoI. The intrinsically bound Mg2+ coordinatesthrough the MstopoI E112 residue (Fig. 10). Theinteratomic coordination distances in the modelprovide an explanation for the retention of Mg2+ inthe MstopoI D108A mutant and its escape fromMstopoI E112A. It appears that the bound Mg2+ inMstopoI E112A (EctopoI E115A) can easily escape tothe surrounding due to the alterations in the pocketand weakened interactions by the other two acidicgroups of the triad. As a result, Mg2+ is no longerefficiently bound to the pocket in MstopoI E112A.Equilibrium dialysis studies with EctopoI E115Arevealed reduced metal ion binding stoichiometrywith amole ratio near oneMg2+ per enzymemoleculeinstead of two normally found in EctopoI.13 As can beseen in Fig. 10, E115A mutation in EctopoI could alsofacilitate release of Mg2+ from the toprim motif.Although ATP-independent type IA and ATP-

dependent type II topoisomerases share a very similarcatalytic center and Mg2+ ion coordination throughthe acidic triad of toprim motif,44,45 subtle variationsare seen with respect to the metal ion coordinationand the role of individual residues during the twosequential transesterification reactions. The presentstudies unambiguously demonstrate that Mg2+ playsno role in DNA cleavage but is required for religationand overall relaxation reaction. These observationsindicate that Mg2+ plays a role after DNA cleavage.Mg2+ may be important in the activation of thenucleophile for religation. As religation step is closelylinked with post-cleavage strand passage and con-formational changes, the current experiments or theearlier data would not be adequate to pinpoint theprecise step of Mg2+ function. While the presentstudies reveal the importance of the first acidic residuein type IA enzymes, a mutation in the third residue ofthe acidic triad,D530A, resulted in severe loss ofDNArelaxation and cleavage activities of S. cerevisiaetopoisomerase II.30 Interestingly, the third acidicresidue in the triad of the toprim domain appears tobe more conserved in the enzymes that carry out thesecond transesterification reaction.9 Even though

phylogenetically conserved, there are subtle differ-ences in the role of the individual residues of thetoprim domain in various topoisomerases. Thetoprim domain in a given enzyme seems to besuitably tailored in its Mg2+ coordination to optimizeits in vivo enzyme function. The key to the optimumcleavage–religation reaction appears to be properalignment of Mg2+ ions with respect to the catalyticresidues and the substrate.In conclusion, the present studies highlight the

importance of toprim domain and its remarkablefunctional conservation in type IA topoisomerases.With a seemingly minor perturbation in the activecenter, the metal ion seems to participate in theactivation of nucleophile in a type IA topoisomerasesimilar to type II enzymes and nucleases. Theexquisite sensitivity of Mg2+ to the local environmentseems to provide a means for the enzyme to convertsmall conformational differences into large catalyticchanges. The minor architectural differences in thetoprim–active site juxtaposition thus appears to play amajor role in diversifying into different functions seenin a variety of toprim possessing DNA/RNAtransaction enzymes.

Materials and Methods

Bacterial strains and plasmids

E. coliDH10B andE. coliBL21 (DE3)wereused for cloningand overexpression of topoisomerase I and itsmutants frompRSETA constructs, respectively. E. coli BL21 (DE3)harboring either pLysS or pLysE plasmid was used forregulated overexpression of MstopoI D108A. SupercoiledpUC18 plasmid DNA substrate for topoisomerase I activityassay was purified by Qiagen midiprep kits.

Enzymes and oligonucleotides

Restriction enzymes, T4 DNA ligase, and T4 polynucle-otide kinase were purchased from New England Biolabs.Oligonucleotides (Sigma andMicrosynth) were resolved on15% denaturing polyacrylamide gel, purified using NAP 10columns, and 5′ end labeled using T4 polynucleotide kinaseand [γ-32P]ATP (6000 Ci/mmol). Labeled oligonucleotideswere purified through Sephadex G-25 spin columns.

Multiple sequence alignment

Amino acid sequences of type IA topoisomerases wereobtained from the National Center for BiotechnologyInformation, and multiple sequence alignment was per-formed by ClustalW version 2.0. The aligned sequenceswere analyzed using the GeneDoc program.

Site-directed mutagenesis

Site-directed mutants in the DXDXE motif of MstopoIwere generated by megaprimer inverse PCR method.Expression plasmid pPVN1 encoding amino-terminal re-gion (1–240 aa)wasused as template.Oligonucleotidesweredesigned with respective mutation in forward primers, and

Table 3. Oligonucleotides used for site-directed mutagenesis

Primer Sequence Restriction site

T7 terminator 5′-GCTAGTTATTGTTCAGCGGTGGG-3′D108A 5′-CCTCGCCACTGCCGGCGACCGCG-3′ NaeID110A 5′-CACTGACGGTGCGCGCGAGGGTG-3′ BssHIIE112A 5′-GGTGACCGCGCGGGTGAGGCCATC-3′ —


restriction site was generated when possible. T7 terminatorsequence was used as the reverse primer (Table 3). Thetruncatedmutant forms of pPVN1were subcloned into full-length topoisomerase I (pPVN123) using NdeI/HindIIIrestriction sites. The mutations were confirmed by restric-tion digestion followed by DNA sequencing of the plasmidDNA.

Overexpression and purification of MstopoI and itsmutants

pPVN123 encoding either MstopoI or its mutant formswere transformed into either E. coli BL21 or E. coli BL21pLysS expression strains. Cultures were grown at 37 °C andinduced with 0.3 mM IPTG at A600 nm=0.6 for 3 h. Thecultures were harvested and checked for expression on 8%SDS-PAGE. The proteins were purified by two columnchromatographic steps using heparin Sepharose followedby SP Sepharose as described before.26

DNA relaxation

Negatively supercoiled pUC18 DNA (500 ng) wasincubated with varying amounts of MstopoI and mutantproteins in 20 μl reaction volume at 37 °C for 30 min. Thereaction mixture contained 40 mM Tris–HCl (pH 8.0),20 mMNaCl, 1 mMEDTA, and 5mMMgCl2. The reactionswere terminated with 0.1% SDS and heating at 75 °C for10 min. The reaction products were resolved on 1.2%agarose gel at 1 V/cm and stained with 0.5 μg/ml ethidiumbromide after electrophoresis.

Electrophoretic mobility shift assays

5′-end-labeled 32-mer oligonucleotide (5′-CAGTGAGC-GAGCTTCCGCTTGACATCCCAATA-3′) containing STSfor MstopoI was used as a specific single-stranded DNAsubstrate. The reaction mixtures containing 40 mM Tris–HCl (pH 8.0), 20 mM NaCl, 1 mM EDTA, and 0.1 pmololigonucleotide were incubatedwith the varying concentra-tions ofMstopoI andmutant proteins (1–200 nM) for 15minon ice. The protein–DNA complexes were resolved fromfreeDNAon8%non-denaturingPAGE (30:0.8) at 4 °CusingTBE (Tris–borate–EDTA) as the running buffer. Theradioactivity associated with free and protein-boundoligonucleotides was visualized by phosphorimager(model BAS 1800; Fuji Film), quantitated using ImageGaugesoftware, and analyzed via GraphPad Prism Version 4.0software.

Oligonucleotide cleavage

Two hundred nanomolar of MstopoI and its mutantproteins was incubated with 0.1 pmol of 32-mer STScontaining oligonucleotide in the same buffer as above at37 °C for 30 min to yield a 5′-end-labeled 19-mer cleavedproduct. Reactions were arrested with 45% formamide dye

and heating at 90 °C for 2 min. Samples were resolved on12% denaturing PAGE using 1× TBE as running buffer at300 V. Rate of DNA cleavage by the enzymes was checkedby taking aliquots of reaction products at various timeintervals, followed by analysis and quantitation.

Religation reactions

5′-end-labeled 32-mer (0.2 pmol) STS containing oligo-nucleotide was used as substrate for cleavage by MstopoIandmutant proteins to yielda 5′-end-labeled19-mer and13-mer covalent protein–DNA adduct as products. Thereactions were carried out in the same buffer as above at37 °C for 15 min. To the reaction mixture, 2 pmol of 5′-end-labeled 11-mer (5′-GAGCTTCCGCT-3′) pre-annealed to0.5 pmol of unlabeled 24-mer (5′-TATTGGGATGT-CAAGCGGAAGCTC-3′) was added and incubated atroom temperature for 5 min. Mg2+, at 5- and 10-mM finalconcentrations, was added to the reaction mix andincubated for 15, 30, 45, and 60 min at 37 °C to yield alabeled 24-mer intermolecular religation product. Reactionswere arrested with 45% formamide dye, followed byheating at 90 °C for 2 min, and the products were resolvedon 15% denaturing PAGE. Analysis and quantitation werecarried out using ImageGauge software and GraphPadPrism Version 4.0 software. For quantifying the extent ofreligation, the amount of cleavage in each lane is taken as100%, and percentage religation depicts the amount ofreligation with respect to the cleavage substrate. Forintramolecular religation, the cleavage reaction was carriedout with enzymes using 32-mer oligonucleotide containingSTS, either pre-annealed to 32-mer non-STS oligonucleotideor alone, in the absence of Mg2+ for 15 min. Intramolecularreligationwas initiated by adding 10mMMg2+, followed byincubation for further 15 min. The reaction products wereresolved on 12% denaturing PAGE, and the cleavageproducts were quantified.

Tryptic digestion

MstopoI and its mutant proteins (8 μM)were subjected totryptic digestion at 24 °C for 20 min in the absence orpresence of Mg2+. The reactions were carried out using 5 ngtrypsin in 50mMTris–HCl (pH 8.0)with 0, 5, 10, and 20mMMg2+. Trypsin was inactivated with buffer containingphenylmethylsulfonyl fluoride, and the reaction productswere electrophoresed on 8% SDS-PAGE and visualized bysilver staining.

Atomic absorption spectrometry

MstopoI and its mutant proteins (20 μM) were dialyzedagainst 50mMKPO4 containing 50mMKCl pretreatedwithChelex-25 resin (Sigma). Another duplicate set of proteinswas denatured with 6 M guanidine–HCl at room temper-ature for 2 h and renatured by step dialysis at 4 °C for 24 hwith buffer containing 40 mM Tris (pH 7.4), 50 mM KCl,10% glycerol, and 10 mM EDTA. The renatured protein


samples were subjected to dialysis against Chelex-treated50 mM KPO4 buffer containing 50 mM KCl. The dialyzedbuffer of respective sets was used as blank. Using Mg2+

lamp and standard Mg2+ solutions from 0.2 to 1 ppm, wedid three concurrent readings of Mg2+ in protein samplesvia atomic absorption spectrometer.

Modeling studies

MstopoI protein sequence was submitted to 3D JIGSAW,a protein comparative modeling server. The availablestructure of EctopoI from the Protein Data Bank (1cy0)was used as template to build the model structure ofMstopoI. The structure was then visualized via PyMOLsoftware for molecular allocation of the active centerresidues Y339, D108, D110, and E112 in mycobacterialtopoisomerase I (Y319, D111, D113, and E115 in EctopoI,correspondingly), which are supposed to be involved incoordination of Mg2+ ion(s). The alanine substitutions weregenerated in the model to visualize the possible change inthe interatomic distances that affect Mg2+ coordination.

Acknowledgements

We thank Y. C. Tse-Dinh and D. N. Rao for criticalreading of the manuscript and T. Bhaduri, D. Sikder,P. Jain, and other members of the VN laboratory forthe reagents and discussions. R. J. Deshpande and B.M. Arif from the Departments of Materials Engineer-ing and Soil Mechanics, Indian Institute of Science(IISc), are acknowledged for assistance in AAS. Wethank the Department of Biochemistry, IISc, for CDand fluorescence measurement studies. M. Selvaraj ofthe Department of Molecular Biophysics Unit, IISc, isacknowledged for his assistance in modeling myco-bacterial topoisomerase I. M.N.L. is a recipient ofCouncil of Scientific and Industrial Research SeniorResearch Fellowship, Goverment of India. This workis supported by grants from the Department ofScience and Technology under Fast Track program(to A.G.B.) and the Indian Council of MedicalResearch (V.N.), by an Indo-Danish collaborativegrant, and by a Centre of Excellence grant from theDepartment of Biotechnology (V.N.), Government ofIndia. V.N. is a recipient of J. C. Bose fellowship of theDepartment of Science and Technology, Governmentof India.

Supplementary Data

Supplementary data associated with this articlecan be found, in the online version, at doi:10.1016/j.jmb.2009.08.064

References

1. Champoux, J. J. (2001). DNA topoisomerases: struc-ture, function, and mechanism. Annu. Rev. Biochem.70, 369–413.

2. Forterre, P., Gribaldo, S., Gadelle, D. & Serre, M. C.(2007). Origin and evolution of DNA topoisomerases.Biochimie, 89, 427–446.

3. Tse-Dinh, Y. C. (1998). Bacterial and archeal type Itopoisomerases. Biochim. Biophys. Acta, 1400, 19–27.

4. Viard, T. & de la Tour, C. B. (2007). Type IAtopoisomerases: a simplepuzzle?Biochimie, 89, 456–467.

5. Corbett, K. D. & Berger, J. M. (2004). Structure,molecular mechanisms, and evolutionary relation-ships in DNA topoisomerases. Annu. Rev. Biophys.Biomol. Struct. 33, 95–118.

6. Shuman, S. (1998). Vaccinia virus DNA topoisome-rase: a model eukaryotic type IB enzyme. Biochim.Biophys. Acta, 1400, 321–337.

7. Frohlich, R. F., Andersen, F. F., Westergaard, O.,Andersen, A. H. & Knudsen, B. R. (2004). Regionswithin the N-terminal domain of human topoisome-rase I exert important functions during strand rotationand DNA binding. J. Mol. Biol. 336, 93–103.

8. Lima, C. D., Wang, J. C. & Mondragon, A. (1994).Three-dimensional structure of the 67K N-terminalfragment of E. coli DNA topoisomerase I. Nature, 367,138–146.

9. Aravind, L., Leipe, D. D. & Koonin, E. V. (1998).Toprim—a conserved catalytic domain in type IA andII topoisomerases, DnaG-type primases, OLD familynucleases and RecR proteins. Nucleic Acids Res. 26,4205–4213.

10. Chen, S. J. &Wang, J. C. (1998). Identification of activesite residues in Escherichia coli DNA topoisomerase I.J. Biol. Chem. 273, 6050–6056.

11. Dracheva, S., Koonin, E. V. & Crute, J. J. (1995).Identification of the primase active site of the herpessimplex virus type 1 helicase–primase. J. Biol. Chem.270, 14148–14153.

12. Strack, B., Lessl, M., Calendar, R. & Lanka, E. (1992). Acommon sequence motif, -E-G-Y-A-T-A-, identifiedwithin the primase domains of plasmid-encoded I-and P-type DNA primases and the alpha protein ofthe Escherichia coli satellite phage P4. J. Biol. Chem. 267,13062–13072.

13. Zhu, C. X., Roche, C. J., Papanicolaou, N., DiPietranto-nio, A. & Tse-Dinh, Y. C. (1998). Site-directed mutagen-esis of conserved aspartates, glutamates and argininesin the active site region of Escherichia coli DNAtopoisomerase I. J. Biol. Chem. 273, 8783–8789.

14. Rezacova, P., Borek, D., Moy, S. F., Joachimiak, A. &Otwinowski, Z. (2008). Crystal structure and putativefunction of small Toprim domain-containing proteinfrom Bacillus stearothermophilus. Proteins, 70, 311–319.

15. Beese, L. S. & Steitz, T. A. (1991). Structural basis forthe 3′–5′ exonuclease activity of Escherichia coli DNApolymerase I: a two metal ion mechanism. EMBO J.10, 25–33.

16. Freemont, P. S., Friedman, J. M., Beese, L. S.,Sanderson, M. R. & Steitz, T. A. (1988). Cocrystalstructure of an editing complex of Klenow fragmentwith DNA. Proc. Natl Acad. Sci. USA, 85, 8924–8928.

17. Joyce, C.M.& Steitz, T.A. (1995). Polymerase structuresand function: variations on a theme? J. Bacteriol. 177,6321–6329.

18. Allemand, F., Mathy, N., Brechemier-Baey, D. &Condon, C. (2005). The 5S rRNA maturase, ribonu-cleaseM5, is a Toprim domain familymember.NucleicAcids Res. 33, 4368–4376.

19. Zhu, C. X., Roche, C. J. & Tse-Dinh, Y. C. (1997). Effectof Mg(II) binding on the structure and activity ofEscherichia coli DNA topoisomerase I. J. Biol. Chem.272, 16206–16210.

http://dx.doi.org/10.1016/j.jmb.2009.08.064


20. Sissi, C. & Palumbo, M. (2009). Effects of magnesiumand related divalent metal ions in topoisomerasestructure and function. Nucleic Acids Res. 37, 702–711.

21. Zhu, C. X. & Tse-Dinh, Y. C. (2000). The acidic triadconserved in type IA DNA topoisomerases is requiredfor binding of Mg(II) and subsequent conformationalchange. J. Biol. Chem. 275, 5318–5322.

22. Bhaduri, T., Bagui, T. K., Sikder, D. & Nagaraja, V.(1998). DNA topoisomerase I from Mycobacteriumsmegmatis. An enzyme with distinct features. J. Biol.Chem. 273, 13925–13932.

23. Bhaduri, T., Sikder, D. & Nagaraja, V. (1998).Sequence specific interaction of Mycobacterium smeg-matis topoisomerase I with duplex DNA.Nucleic AcidsRes. 26, 1668–1674.

24. Sikder, D. & Nagaraja, V. (2000). Determination of therecognition sequence of Mycobacterium smegmatistopoisomerase I on mycobacterial genomic sequences.Nucleic Acids Res. 28, 1830–1837.

25. Sikder, D. & Nagaraja, V. (2001). A novel bipartitemode of binding ofM. smegmatis topoisomerase I to itsrecognition sequence. J. Mol. Biol. 312, 347–357.

26. Jain, P. & Nagaraja, V. (2006). Indispensable, func-tionally complementing N- and C-terminal domainsconstitute site-specific topoisomerase I. J. Mol. Biol.357, 1409–1421.

27. Tse-Dinh, Y. C. & Beran-Steed, R. K. (1988). EscherichiacoliDNA topoisomerase I is a zinc metalloprotein withthree repetitive zinc-binding domains. J. Biol. Chem.263, 15857–15859.

28. Studier, F. W. & Moffatt, B. A. (1986). Use ofbacteriophage T7 RNA polymerase to direct selectivehigh-level expression of cloned genes. J. Mol. Biol. 189,113–130.

29. Leelaram,M.N., Bhat, A. G., Suneetha,N., Nagaraja, V.&Manjunath, R. (2009). Immunological cross-reactivityof mycobacterial topoisomerase I and divergence fromother bacteria. Tuberculosis (Edinb.), 89, 256–262.

30. Dong, K. C. & Berger, J. M. (2007). Structural basis forgate-DNA recognition and bending by type IIAtopoisomerases. Nature, 450, 1201–1205.

31. Noble, C. G. &Maxwell, A. (2002). The role of GyrB inthe DNA cleavage–religation reaction of DNA gyrase:a proposed twometal-ion mechanism. J. Mol. Biol. 318,361–371.

32. Sissi, C., Chemello, A., Vazquez, E., Mitchenall, L. A.,Maxwell, A. & Palumbo, M. (2008). DNA gyraserequires DNA for effective two-site coordination ofdivalent metal ions: further insight into the mecha-nism of enzyme action. Biochemistry, 47, 8538–8545.

33. Deweese, J. E., Burgin, A. B. & Osheroff, N. (2008).Human topoisomerase IIalpha uses a two-metal-ionmechanism for DNA cleavage. Nucleic Acids Res. 36,4883–4893.

34. Yang, W. (2008). An equivalent metal ion in one- andtwo-metal-ion catalysis. Nat. Struct. Mol. Biol. 15,1228–1231.

35. Yang, W., Lee, J. Y. & Nowotny, M. (2006). Makingand breaking nucleic acids: two-Mg2+-ion catalysisand substrate specificity. Mol. Cell, 22, 5–13.

36. Nowotny, M., Gaidamakov, S. A., Crouch, R. J. &Yang, W. (2005). Crystal structures of RNase H boundto an RNA/DNA hybrid: substrate specificity andmetal-dependent catalysis. Cell, 121, 1005–1016.

37. Tse-Dinh, Y. C. (1986). Uncoupling of the DNAbreaking and rejoining steps of Escherichia coli type IDNA topoisomerase. Demonstration of an activecovalent protein–DNA complex. J. Biol. Chem. 261,10931–10935.

38. Changela, A., DiGate, R. J. & Mondragon, A. (2007).Structural studies of E. coli topoisomerase III–DNAcomplexes reveal a novel type IA topoisomerase–DNA conformational intermediate. J. Mol. Biol. 368,105–118.

39. Berger, J. M. (1998). Structure of DNA topoisomerases.Biochim. Biophys. Acta, 1400, 3–18.

40. Berger, J. M. (1998). Type II DNA topoisomerases.Curr. Opin. Struct. Biol. 8, 26–32.

41. Depew, R. E., Liu, L. F. & Wang, J. C. (1978).Interaction between DNA and Escherichia coli proteinomega. Formation of a complex between single-stranded DNA and omega protein. J. Biol. Chem. 253,511–518.

42. Cheng, B., Annamalai, T., Sorokin, E., Abrenica, M.,Aedo, S. & Tse-Dinh, Y. C. (2009). Asp-to-Asnsubstitution at the first position of the DxD TOPRIMmotif of recombinant bacterial topoisomerase I isextremely lethal to E. coli. J. Mol. Biol. 385, 558–567.

43. Perry, K. & Mondragon, A. (2002). Biochemicalcharacterization of an invariant histidine involved inEscherichia coli DNA topoisomerase I catalysis. J. Biol.Chem. 277, 13237–13245.

44. Berger, J. M., Fass, D., Wang, J. C. & Harrison, S. C.(1998). Structural similarities between topoisomerasesthat cleave one or both DNA strands. Proc. Natl Acad.Sci. USA, 95, 7876–7881.

45. Liu, Q. & Wang, J. C. (1999). Similarity in the catalysisof DNA breakage and rejoining by type IA and IIADNA topoisomerases. Proc. Natl Acad. Sci. USA, 96,881–886.

Deciphering the Distinct Role for the Metal Coordination Motif in the Catalytic Activity of...

Documents

Transcript of Deciphering the Distinct Role for the Metal Coordination Motif in the Catalytic Activity of...