| Nucleic Acids Research | Pages |
Step-wise DNA relaxation and decatenation by NaeI-43K
Introduction
Materials And Methods
DNA relaxation assay
Ligation of nicked pBR322 DNA circles
Isolation of a unique pBR322 topoisomer
Decatenation reaction
Results And Discussion
NaeI-43K DNA relaxation uses a single step concerted mechanism
NaeI is a type I topoisomerase
Processive and distributive modes of action
Kinetics of DNA relaxation
Decatenation
Conclusions
Acknowledgements
References
Step-wise DNA relaxation and decatenation by NaeI-43K
ABSTRACT
INTRODUCTION
NaeI endonuclease from Nocardia aerocolonigenes either sequence-specifically cleaves double-stranded DNA or relaxes supercoiled DNA, depending upon whether position 43 is leucine (NaeI-43L) or lysine (NaeI-43K) respectively (1). DNA relaxation activity appears to result from coupling of ligase and endonuclease activities by means of a transient covalent intermediate that conserves the energy released during DNA cleavage for use during religation. Thus restriction endonuclease NaeI is related and probably evolved from a topoisomerase/ligase (1,2). Whereas endonuclease NaeI-43L binds double-stranded DNA with sequence specificity and a Kd of ~30 nM (2-5), topoisomerase NaeI-43K binds DNA with relaxed specificity, binding DNAs without a cognate recognition sequence, and preferentially binds single-stranded and mismatched DNA with a Kd of ~1.2 µM (2). The low binding affinity of NaeI-43K means that high amounts of protein are needed to achieve significant activity. The preference for binding single-stranded DNA suggests that NaeI-43K is related to the IA subfamily of DNA topoisomerases, like DNA topoisomerase I from Escherichia coli and DNA topoisomerase III from yeast and E.coli (for a description of the IA and IB subfamilies see ref. 6).
One important question that relates to the mechanism by which NaeI-43K relaxes DNA is whether it changes linking number by units of one or two. Since NaeI-43K is derived from a type II endonuclease that cleaves double-stranded but not single-stranded DNA, it is natural to suspect that NaeI-43K topoisomerase activity is type II. However, both NaeI-43L (5) and NaeI-43K (1,2) generate nicked intermediates during DNA cleavage, depending upon conditions, so that type I topoisomerase activity is also possible for NaeI-43K.
Another important question is whether NaeI-43K can interact simultaneously with both newly cleaved DNA ends to control strand passage and to relax DNA in a stepwise `concerted' fashion. (We use concerted and unconcerted here to differentiate between cleavage, strand passage and ligation that are unified to change linking number in discrete units and random coupling that leads to random or large changes in linking number.) This question arises because NaeI-43K topoisomerase activity apparently derives from the coupling of endonuclease and ligase activities. This coupling of two opposing activities in principle makes it possible for many strand passages to occur during a cleavage-rejoining cycle that is terminated by a more-or-less random religation event (7).
To answer these questions we isolated a single topoisomer of pBR322 and studied the ability of NaeI-43K to change its linking number. We also studied the effects of salt concentration on the ability of NaeI-43K to relax DNA and we investigated the ability of NaeI-43K to decatenate kinetoplast DNA (k-DNA) and characterized the products of this reaction. The results of these experiments demonstrate that NaeI-43K changes DNA linking number in units of one rather than two, relaxes supercoiled DNA in an apparently step-wise concerted manner and decatenates interlocked nicked DNA circles.
MATERIALS AND METHODS
DNA relaxation assay
The standard relaxation assay (15 µl) contained 10 mM Tris-HCl, pH 8.0, 20 mM NaCl, 5 mM MgCl2, 0.1 mg/ml bovine serum albumin and 5.0 mM [beta]-mercaptoethanol. Supercoiled pBR322 DNA and purified NaeI-43K were incubated at 37°C and stopped by addition of SDS to a final concentration of 1%. Products were resolved by 0.8% agarose gel electrophoresis in TAE buffer (40 mM Tris-acetate, pH 8.0, and 1 mM EDTA) and stained with 0.5 µg/ml ethidium bromide (EtBr).
Ligation of nicked pBR322 DNA circles
Supercoiled pBR322 DNA was nicked by limited digestion with DNase I (8 ng/ml) and purified by extraction with phenol and chloroform, followed by ethanol precipitation. Equilibrium topoisomer distributions were produced by ligating the nicked pBR322 DNA (11.6 nM) with T4 DNA ligase (3 Weiss U/ml) in 15 ml reactions containing 10 mM Tris-HCl, pH 8.0, 5 mM MgCl2, 0.5 mM ATP, 0.1 mg/ml bovine serum albumin, 5.0 mM [beta]-mercaptoethanol and indicated amounts of NaCl. The reactions were incubated at 37°C for 3 h and stopped by addition of SDS to a final concentration of 1%. Topoisomer products were resolved by 0.8% agarose gel electrophoresis in TAE buffer and visualized by staining with 0.5 mg/ml EtBr.
Isolation of a unique pBR322 topoisomer
Plasmid pBR322 (200 µg) was partially relaxed with 50 U/ml Drosophila topoisomerase II (US Biochemical) using the reaction conditions indicated by the provider. The reaction products were electrophoresed for 24 h at 2 V/cm on a preparative 0.8% agarose gel. The gel was stained for 15 min with 0.5 µg/ml EtBr, the DNA bands of each topoisomer were visualized with long wavelength UV light and the bands were individually excised from the gel. The DNA in each slice was eluted by electrophoresis into the TAE electrophoresis buffer in a dialysis bag and then precipitated with ethanol.
Decatenation reaction
k-DNA (400 µg) (TopoGEN Inc.) and NaeI-43K (0.72 µM) were incubated at 37°C for 1 h under the reaction conditions as described in DNA relaxation assay. Reaction products were analyzed by 0.8% agarose gel electrophoresis and quantitated by densitometry after staining with EtBr.
RESULTS AND DISCUSSION
NaeI-43K DNA relaxation uses a single step concerted mechanism
NaeI-43K forms topoisomers of supercoiled DNA (1), forms a transient covalent intermediate with the 5[prime]-end of the newly cleaved strand (1,8) and binds both single-stranded and double-stranded DNA, with a preference for the former (2). These characteristics are similar to those of prokaryotic type I topoisomerases (6,9-15). In principle, topoisomerase I can remove from one (concerted) to many supercoils from a DNA molecule during one cleavage-religation cycle, depending on the ability of the enzyme to bridge the two cleaved ends (discussed in 6,9-13). Strong interaction between topoisomerase and the newly cleaved DNA ends is necessary to tightly control the number of supercoils lost during one cleavage-rejoining cycle. The loss of many supercoils during one cycle is a possibility for NaeI-L43K because its activity may be derived from a loose coupling of endonuclease and ligase domains. In the complete absence of enzyme bridging of the cleaved DNA ends the relaxation reaction will proceed by a random rather than step-wise mechanism to give an equilibrium Boltzmann distribution of topoisomers (7). This equilibrium distribution is generated when DNA nicked by DNase I is closed by DNA ligase and the products resolved by gel electrophoresis (7,16; Fig. 1, lanes 3, 5 and 7). NaeI-43K under conditions identical to the ligase reaction reduced the distribution of topoisomers and drove the reaction to near completion at all salt concentrations tested up to 100 mM NaCl (Fig. 1, lanes 4, 6 and 8).
Figure 1. Ability of NaeI-43K to relax supercoiled DNA. Ligation of nicked pBR322 (11.6 nM) using T4 DNA ligase (3 Weiss U/ml) and DNA relaxation of supercoiled pBR322 (11.6 nM) using NaeI-L43K (0.25 µM) were performed under identical conditions as described in Materials and Methods (ligation of nicked pBR322 circles). Lane 1, DNase I-nicked pBR322; lane 2, supercoiled pBR322; lanes 3, 5 and 7, nicked pBR322 incubated with T4 DNA ligase at 10, 60 and 100 mM NaCl; lanes 4, 6 and 8, pBR322 incubated with NaeI-43K at 10, 60 and 100 mM NaCl. Reaction products were resolved by 0.8% agarose gel electrophoresis and stained with EtBr. The DNA in the bands indicated as relaxed and nicked in Figure 1, lanes 4, 6 and 8 was shown to be composed of 85% covalently closed circles and 15% nicked circles by electrophoresing the reaction products under identical conditions except for addition of EtBr (not shown). The large amount of covalently closed circles indicates the ability of high concentrations of NaeI-43K to drive the reaction to near completion (Fig. 1, lanes 3, 5 and 7 versus lanes 4, 6 and 8 respectively). Lower concentrations of NaeI-43K and long incubation times should give the equilibrium distribution of topoisomers. Lower concentrations of NaeI-43K could not be tested, however, because the enzyme is not stable for more than 1 h under these conditions. The narrowed distribution of topoisomers produced by NaeI-43K at high concentration (Fig. 1) was neither the result of protein binding to intact topoisomers during electrophoresis nor the result of nicked molecules with bound protein being converted to closed molecules upon analysis in EtBr gels: treatment with proteinase K under conditions found to inactivate NaeI-43K had no effect either on the distribution of topoisomers observed after gel electrophoresis or on the amount of covalently closed products resolved by EtBr gel electrophoresis (not shown). Since NaeI does not use ATP or NAD to drive the reaction below equilibrium, as does topoisomerase II (17), NaeI must interact with the DNA in a way that distorts the equilibrium. How it does this should be important for understanding NaeI-43K-DNA interactions. The ratio of NaeI-43K to DNA used in the reactions shown if Figure 1, ~11 NaeI dimer molecules for each DNA molecule, argues against formation of a protein-DNA filament. To investigate this possibility further, NaeI in the relaxation reaction was fixed by addition of glutaraldehyde and the products visualized by electron microscopy as described elsewhere (18). No large protein complexes were found over a significant distribution of topoisomers (not shown). One alternative possibility for NaeI-43K-DNA interaction comes from the ability of NaeI to juxtapose distant DNA sequences to give crossover-like complexes (18). Perhaps formation of several of these complexes on a DNA molecule by NaeI-43K, which has relaxed recognition specificity relative to NaeI (2), narrows the distribution below equilibrium by partitioning the DNA into several relaxed and supercoiled regions. The concentration of supercoils into a smaller space allows NaeI to achieve a net lowering of the topoisomer distribution below equilibrium. Another possibility relates to the preference of NaeI-43K for binding single-stranded DNA (2). Significant denaturation of regions of the DNA would also reduce the amount of supercoiling.
NaeI is a type I topoisomerase
Topoisomerases can change DNA linking number in units of one (type I) or two (type II). To demonstrate that NaeI-43K has type I activity we isolated a single negatively supercoiled topoisomer from pBR322 partially relaxed by Drosophila topoisomerase II. Products of the relaxation of that topoisomer by topoisomerase II and by NaeI-43K were analyzed on agarose gels that resolve topoisomers differing in linking number by units of one (19,20). The results showed clearly that NaeI-43K acts like topoisomerase I in its ability to change the DNA linking number in units of one (Fig. 2, lanes 3-5) rather than two, as shown in Figure 2, lane 6 for Drosophila topoisomerase II (21). Another single topoisomer of pBR322 was isolated by cleavage of pBR322 and religation at 4°C. A single positively supercoiled topoisomer was isolated from an agarose gel electrophoresed at room temperature (16). This topoisomer was resistant to NaeI-43K (not shown), demonstrating that NaeI-43K only relaxes negative supercoils, as expected for an enzyme that recognizes single-stranded DNA (2), and thus must rely on negative supercoiling to help generate single-stranded regions in the DNA substrate.
Figure 2. NaeI-43K changes linking number of a unique pBR322 topoisomer by units of one. DNA relaxation reactions were performed by incubating NaeI-43K and Drosophila topoisomerase II with an isolated pBR322 topoisomer of unique linking number (11.6 nM) under the conditions described in Materials and Methods. Reaction products were resolved by 0.8% agarose gel electrophoresis and stained with EtBr. Lane 1, pBR322 linearized with HindIII digestion; lane 2, unique linking number topoisomer only; lanes 3-5, topoisomers generated by NaeI-43K (0.13 µM) with incubation times of 2, 5 and 20 min respectively; lane 6, topoisomers generated by Drosophila topoisomerase II. Changes in linking number ([Delta]Lk) are indicated.
Processive and distributive modes of action
To assess how actively NaeI-43K controls the loss of supercoils during one cleavage-rejoining cycle we determined the effects of [Na+] and [Mg2+] on DNA relaxation by NaeI-43K using gel electrophoresis (Figs 3 and 4). At low salt concentrations (0-30 mM NaCl) the action of NaeI-43K was at least partially processive. The concentration of NaeI-43K was ~5-fold below its Kd for single-stranded DNA of ~1.2 µM (2), so that the reaction with single-stranded regions in pBR322 were not saturating. Under these conditions, reaction of NaeI-43K with pBR322 left a fraction of the substrate fully supercoiled (Fig. 3, lanes 2 and 3); a bimodal distribution of supercoiled and relaxed molecules with a distribution or ladder of topoisomers in between that approximated the equilibrium distribution (Fig. 1) was evident. At higher NaCl concentrations (50-110 mM) the action of the same non-saturating amount of NaeI-43K was distributive (Fig. 3, lanes 4-6 and 9). All of the pBR322 molecules were affected. The distribution of the resulting topoisomers under the higher salt conditions was apparently Gaussian, with no fully supercoiled molecules remaining. The distance of the distribution mean from the gel origin and the size of the deviation about the mean decreased with increasing salt concentration from 30 to 110 mM NaCl. The reaction appeared to be completely inhibited at [ge]160 mM NaCl (Fig. 3, lanes 8 and 9), although the gels may hide some activity by not resolving molecules with small changes in linking number.
Figure 3. Effect of NaCl on NaeI-43K relaxation of pBR322. NaeI-43K (0.25 µM) was incubated with pBR322 (23 nM) for 1 h under the conditions described in Materials and Methods, except with increasing concentrations of NaCl. Lane 1, pBR322 only; lanes 2-9, NaeI-43K relaxation reaction with 10, 30, 50, 70, 90, 110, 160 and 210 mM NaCl respectively. Figure 4. Effect of MgCl2 on NaeI-43K relaxation of pBR322. NaeI-43K (0.25 µM) was incubated with pBR322 (23 nM) for 1 h under the conditions described in Materials and Methods except with increasing concentration of MgCl2. Lanes 1-8, NaeI-43K relaxation reaction with 0, 1, 2, 4, 6, 8, 16 and 32 mM MgCl2. The change from processive to distributive mode of action with increasing salt concentration is generally seen for DNA binding proteins when the electrostatic interaction component contributing to the processiveness of the interaction is overcome by Debye-Hückle shielding (for a definition see 22). As predicted for such an effect, the divalent cation Mg2+ was significantly more effective at both switching the distribution of products from bimodal ([le]6 mM MgCl2, Fig. 4, lanes 1-5) to Gaussian (>6 mM MgCl2, lanes 6 and 7) and inhibiting the reaction (32 mM MgCl2, lane 8).
Kinetics of DNA relaxation
DNA relaxation by NaeI-43K under processive, distributive and intermediate conditions was studied with increasing time at different salt concentrations determined from Figure 4. At low (10 mM) salt concentration the reaction was too fast to clearly observe formation of intermediate topoisomers (Fig. 5). The reaction clearly gave a bimodal distribution of supercoiled and relaxed molecules. At 60 mM NaCl the intermediate topoisomers became more discernible. The reaction under these conditions was partially processive, as seen by the mostly bimodal distribution observable at early times (lanes 2 and 3), and partially distributive, as seen by the complete loss of fully supercoiled substrate and concomitant Gaussian distribution of topoisomers at later times (lanes 4 and 5). At 100 mM NaCl, topoisomer band formation was clearly observed and the Gaussian distribution of topoisomer intermediates, characteristic of distributive behavior, proceeded from fully supercoiled towards the fully relaxed state in a step-wise fashion (lanes 1-6). These results imply that the number of supercoils lost in one DNA cleavage-religation cycle is, as it is for E.coli topoisomerase I (14), close to one. A low number of supercoils lost per DNA cleavage-rejoining cycle implies significant interaction of NaeI-43K with both newly cleaved DNA ends. Such interaction is required to achieve concerted step-wise DNA relaxation.
Figure 5. Effects of NaCl on the kinetics of pBR322 relaxation by NaeI-43K. NaeI-43K (0.13 µM) was incubated with pBR322 DNA (11.6 nM) for the indicated time in the presence of 100 (top), 60 (middle) or 10 (bottom) mM NaCl. Reaction conditions are described in Materials and Methods.
Decatenation
The ability to catenate and decatenate DNA is characteristic of the topoisomerases (6,9-13). Type I topoisomerases require that the DNA circle to be decatenated be nicked (23-25), whereas type II topoisomerases do not. The product of topoisomerase I decatenation is nicked circles. k-DNA from the trypanosomatid Crithidia fasciculata is composed of a network of interlocked rings of two different sizes. The network can be isolated either as Form I, in which all circles of the network are covalently closed, or as the replicative intermediate Form II, in which all the circles of the network are nicked (reviewed in 26). Incubation of Form I k-DNA with Drosophila topoisomerase II and NaeI-43K showed that topoisomerase II but not NaeI-43K could decatenate the k-DNA to give covalently closed circles (not shown). On the other hand, Form II k-DNA could be decatenated by both enzymes (Fig. 6, compare lanes 3 and 6 with lane 2). m-AMSA, a eukaryotic topoisomerase II inhibitor that inhibits the ability of NaeI-43K to form a covalent intermediate (8), inhibited NaeI-43K decatenation of k-DNA by 40% at 100 µM drug, as measured by densitometry of the gel bands (lanes 3-5). Using Form II k-DNA as substrate NaeI-43K gave the identical mobility product to that of topoisomerase II. The decatenation products could be distinguished from linear DNA by their electrophoretic mobility in agarose gels (Fig. 6, compare lanes 2-6 with lanes 1 and 7). Linear DNA is the product of cleaving the circles to take them apart, whereas the decatenation products of NaeI-43K and of topoisomerase II are separated DNA circles. Similar results to that in Figure 6 were obtained under distributive salt conditions of 100 mM NaCl (not shown).
Figure 6. NaeI-43K decatenates k-DNA. Lanes 1 and 7, k-DNA linearized by XhoI digestion; lane 2, catenated k-DNA only; lanes 3-5, k-DNA incubated with NaeI-43K in the presence of 0, 50 and 100 mM m-AMSA respectively; lane 6, k-DNA incubated with Drosophila topoisomerase II. Reaction conditions are described in Materials and Methods and products are shown resolved by 0.8% agarose gel electrophoresis and stained with EtBr. The products of the decatenation reactions were examined further on an agarose gel containing EtBr to distinguish between relaxed covalently closed and nicked circular molecules (Fig. 7). The topoisomerase II catalyzed reaction products were mainly composed of nicked circular DNA. A small amount of product was covalently closed, as shown by its increased mobility in the presence of EtBr (lane 2). The only product in the decatenation reaction catalyzed by NaeI-43K was nicked circular DNA (lane 3). There was a small amount of k-DNA remaining after the reaction had gone to completion. We assume that the remaining k-DNA was the amount of covalently closed catenated DNA converted to closed circles by topoisomerase II in lane 2. These results imply that, like topoisomerase I, NaeI-43K can cleave opposite a nick to decatenate nicked catenanes. Figure 7. NaeI-43K requires nicked double-stranded DNA circles for decatenation. The products of the decatenation reactions described in Figure 6 were resolved on a 1% agarose gel containing EtBr (0.5 µg/ml) to differentiate nicked circular from closed circular k-DNA. Lane 1, k-DNA only; lane 2, k-DNA incubated with Drosophila topoisomerase II; lane 3, k-DNA incubated with NaeI-43K.
Conclusions
Our results demonstrate that NaeI-43K is a type I topoisomerase that relaxes only negatively supercoiled DNA and imply that NaeI-43K successfully bridges the two cleaved ends to relax DNA in a step-wise concerted manner. Changing salt concentration switched the mode of action of NaeI-43K from processive to distributive and the kinetics of the reaction imply a low number of supercoils lost in each DNA molecule during one cleavage-religation cycle. NaeI was also shown to decatenate k-DNA that contained nicked circles. The latter result, together with the known preference of NaeI-43K for binding single-stranded and mismatched DNA (2), implies that NaeI-43K can cleave opposite a nick. These results strengthen the relationship between NaeI and the topoisomer/ligase family of proteins. They imply that the ligase and endonuclease activities are strongly coupled both by use of a covalent intermediate and by enzyme bridging of both cleaved ends in order to control DNA strand passage.
ACKNOWLEDGEMENTS
We thank Jack Griffith for electron microscopy. This work was supported by grant GM52123 from the National Institutes of Health.
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 13 May 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
K. L. Carrick and M. D. Topal
Amino Acid Substitutions at Position 43 of NaeI Endonuclease. EVIDENCE FOR CHANGES IN NaeI STRUCTURE
J. Biol. Chem.,
March 7, 2003;
278(11):
9733 - 9739.
[Abstract]
[Full Text]
[PDF]
J. Bitinaite and I. Schildkraut
Self-generated DNA termini relax the specificity of SgrAI restriction endonuclease
PNAS,
January 24, 2002;
(2002)
22346799.
[Abstract]
[Full Text]
[PDF]
J. Bitinaite and I. Schildkraut
Self-generated DNA termini relax the specificity of SgrAI restriction endonuclease
PNAS,
February 5, 2002;
99(3):
1164 - 1169.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (320K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (5)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Jo, K.
Articles by Topal, M. D.
Search for Related Content
PubMed
PubMed Citation
Articles by Jo, K.
Articles by Topal, M. D.
Social Bookmarking
What's this?