Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (136K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Zinkevich, V.
Right arrow Articles by Firman, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zinkevich, V.
Right arrow Articles by Firman, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1997 Oxford University Press 503-511

Footnote

The HsdR subunit of R . Eco R124II: cloning and over-expression of the gene and unexpected properties of the subunit

The HsdR subunit of R . Eco R124II: cloning and over-expression of the gene and unexpected properties of the subunit Vitaly Zinkevich + , Ljuba Popova 1 , Valentin Kryukov 1 , Agnes Abadjieva , Irina Bogdarina 1 , Pavel Janscak and Keith Firman*

Biophysics Laboratories, School of Biological Sciences, University of Portsmouth, St Michael's Building, White Swan Road, Portsmouth PO1 2DT, UK and 1 Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Pushchino on the Oka , 142292, Russia

Received October 24, 1996; Revised and Accepted December 9, 1996

ABSTRACT

Type I restriction endonucleases are composed of three subunits, HsdR, HsdM and HsdS. The HsdR subunit is absolutely required for restriction activity; while an independent methylase is composed of HsdM and HsdS subunits. DNA cleavage is associated with a powerful ATPase activity during which DNA is translocated by the enzyme prior to cleavage. The presence of a Walker type I ATP-binding site within the HsdR subunit suggested that the subunit may be capable of independent enzymatic activity. Therefore, we have, for the first time, cloned and over-expressed the hsdR gene of the type IC restriction endonuclease Eco R124II. The purified HsdR subunit was found to be a soluble monomeric protein capable of DNA- and Mg 2+ -dependent ATP hydrolysis. The subunit was found to have a weak nuclease activity both in vivo and in vitro , and to bind plasmid DNA; although was not capable of binding a DNA oligoduplex. We were also able to reconstitute the fully active endonuclease from purified M [middot] Eco R124I and HsdR. This is the first clear demonstration that the HsdR subunit of a type I restriction endonuclease is capable of independent enzyme activity, and suggests a mechanism for the evolution of the endonuclease from the independent methylase.

INTRODUCTION

Type I restriction endonucleases are multimeric enzymes composed of three different subunits, with complex modes of action: they function as an endonuclease, a methylase, an ATPase and can translocate DNA ( 1 , 2 ). The HsdS subunit is required for DNA recognition ( 3 ) and together with HsdM comprises an independent DNA methyltransferase ( 4 , 5 ); the additional subunit required for production of the endonuclease is the HsdR subunit. Several conserved sequences have been identified within the HsdR subunit including a region, found in many helicases, which may be required for ATP-dependent DNA translocation ( 6 ).

For the best characterised type I restriction endonuclease Eco KI, it is known that restriction activity is dependent on the cofactors S -adenosyl methionine (SAM), ATP and Mg 2+ . SAM is required early in the reaction pathway as an allosteric effector which transforms the enzyme into a form which binds at the recognition site. In the presence of ATP, the state of methylation of the recognition sequence determines whether DNA cleavage, or DNA methylation occurs ( 7 , 8 ).

Double-strand DNA cleavage has been proposed to involve two endonuclease molecules, which act co-operatively when they meet following DNA translocation ( 9 ). The DNA translocation is driven by the ATPase activity observed during this stage ( 10 , 11 ). The endonuclease has no turnover ( 8 ), and appears to be altered following cleavage. This altered enzyme behaves as a powerful ATPase ( 11 ). The lack of turnover is thought to reflect the enzyme being covalently bound to the DNA following cleavage ( 10 , 11 ).

The Eco R124II R-M system is a member of the type IC family of I restriction enzymes and like all other type I enzymes recognises a `split' DNA sequence. The other members of this group are listed below: GAAnnnnnnRTCG Eco R124I ( 12 ) GAAnnnnnnnRTCG Eco R124II (formally called Eco R124/3) TCAnnnnnnnATTC Eco DXXI ( 13 ) CCAnnnnnnnRTGC Eco prrI ( 14 )

The only difference between the recognition sequences of the Eco R124I and Eco R124II R-M systems is the presence of an extra nucleotide in the `spacer' region of the s R124II recognition sequence. This difference is mirrored by a single change in the HsdS subunit. The HsdS subunit from the Eco R124II R-M system contains an extra repeat of four amino acids within the central conserved region ( 15 ). Since there are no differences between the hsdR genes of Eco R124I and Eco R124II, the HsdR subunit isolated from one system will assemble with the subunits from the other system.

Table 1 . Bacterial strains, bacteriophage and plasmids

Genotype or phenotype

Source

Bacterial strain

C122

prototroph

British Culture Collection

Strain No. 122

2K

thi, serB, lac, leu r - K m - K str-r

S.W.Glover

JM109

F' tra [Delta] 36, lacI, [Delta]( lacZ )M15 proAB / recA1 , endA1, gyrA96 (Nal R ),

(30)

hsdR17, mcrA, relA!, supE, sbcBC , thi-1, [Delta]( lac-proAB )

JM109(DE3)

As above, but with [lambda] ([Delta]E3).

Promega

Bacteriophage

[lambda] vir

(31)

Plasmids

pUNG31

pBR322 carrying the hsd genes from R124/3

(21)

pTZ18/19R

Ap R plasmid carrying the T7 g10 promoter

(32)

pDRM-1R

Single s R124I site [duplex I from Taylor et al . (4)]

Darren Mernagh

cloned into pTZ18R

R124/3

Native plasmid carrying the hsd genes of Eco R124II

(33)

pAC3MS

pACYC184 carrying the hsdM and hsdS genes from pUNG31

This paper

expressed from their natural promoter

pLP17 & 25

pTZ19R carrying the hsdR gene from pUNG31, expressed from

This paper

its natural promoter, in both orientations respectively

pETR124

pET3A carrying the hsdR gene expressed from the

This paper

T7 g10 promoter

pTrc99A

Ap R , lacI q , P trc

(22)

pVZR2

pTrc99A carrying the hsdR gene expressed from the

This paper

P trc promoter

pJS4M

A derivative of pUC119 and pET3a over-producing M. Eco R124I

(34)

methylase from T7 promoters

A number of groups have worked on type I R-M systems for several years, and it was not expected that these systems would provide new information. However, recently a number of startling discoveries have been made with these enzymes. In particular, novel DNA specificities have been isolated following deletion mutagenesis ( 16 , 17 ) and interaction with another subunit have resulted in production of an anti-codon nuclease ( 18 ). In this paper we describe the use of a high-level expression system to over-express the hsdR gene from the trc promoter. We show an ATPase activity of the purified HsdR subunit and, surprisingly, additional activities for the subunit as an in vivo and in vitro nuclease and the ability to bind DNA. In addition, we show that we can reconstitute an active restriction endonuclease using this HsdR subunit.

MATERIALS AND METHODS

Bacterial strains, bacteriophage and plasmids used in this study are listed in Table 1 . In vivo tests for restriction and modification were as previously described ( 3 ) and were based on the determination of the efficiency of plating of bacteriophage lambda. Plasmid DNA isolation, and manipulation in vitro , involved standard techniques ( 19 ). Restriction enzymes and DNA modifying enzymes were used according to the manufacturer's recommendations.

Construction of recombinant plasmids

Construction of the plasmid pAC3MS, a plasmid compatible with pBR-based vectors and capable of producing M [middot] Eco R124II methylase for use in complementation analysis, was accomplished by digestion of pUNG31 with Ple 19I (an isoschizomer of Pvu I), followed by incubation with T4 DNA polymerase to produce blunt-ends. This DNA was then digested with Hin dIII and the 4.2 kb fragment isolated from a preparative agarose gel. This DNA was ligated to a Hin dIII + Eco RV digestion of pACYC184 ( 20 ) and used to transform Escherichia coli C122. Recombinant clones were detected as Cm R Tc S on antibiotic indicator plates.

A computer search of the Eco R124II hsd DNA sequence ( 15 ) using the DNASTAR MAPDRAW program demonstrated that the hsdR gene was entirely contained within a single Bsa I DNA fragment. This fragment also carries the C-terminal 856 bp of the hsdS gene. The first step employed in the construction of an over-producing plasmid carrying hsdR was to ligate the Bsa I fragment isolated from pUNG31 ( 21 ) to a Sma I digest of pTZ19R, following conversion of the Bsa I DNA ends to blunt-ends using Klenow fragment of E.coli DNA polymerase I and dNTPs. Transformation of JM109 yielded two types of recombinant plasmid, identified by screening the transformants for the presence of an asymmetrical Age I restriction site. These plasmids were named pLP17 and pLP25 and have the opposite orientation to the inserted fragment.

To obtain over-production of HsdR, using the plasmid pET3a, it was necessary to introduce a Nde I restriction site overlapping the ATG start codon of the hsdR gene. This would allow fusion of the gene to the Shine-Dalgarno sequence of pET3a. An oligonucleotide, which introduces a Nde I by means of a single base insertion, was used as one PCR primer (see primer 1, PCR section); the other primer was designed to overlap a unique Bal I within the hsd R gene (primer 2, PCR section). PCR using these two primers and the pLP25 plasmid DNA as a template resulted in production of a 200 bp product (Fig. 1 ). Cleavage of this PCR product with Bal I + Nde I yielded a 180 bp fragment carrying the N-terminal region of hsdR . Plasmid pLP25 was cut with Bal I + Bam HI to release the remaining part of the hsdR gene. This fragment and the Bal I/ Nde I digested PCR fragment were ligated to the large Nde I- Bam HI fragment of pET3a (purified from a low-melting agarose gel), and the resulting ligation products used to transform JM109. Recombinants were identified as containing the required DNA fragment following digestion with Nde I + Bam HI; the resulting plasmid was named pETR124.


Figure 1 . Construction of plasmids used in this work. The hsdR gene was initially isolated in pTZ19R as a single Bsa I fragment, and confirmed as such by complementation to Res + with pAC3MS. The resultant plasmid pLP17 was used as a template for PCR to produce the small PCR-product illustrated (which was designed to incorporate a Nde I site at the start codon). The small PCR product was ligated to the large Bal I- Bsa I fragment of pLP17 to give an intact hsdR gene carrying a Nde I site, which was subsequently inserted between the Nde I and Bam HI sites of pET3a, or the Nco I- Bam HI sites of pTrc99A.

Construction of pVZR2 was accomplished following ligation of the Nde I- Hin dIII fragment from pETR124 to the large Nco I + Hin dIII fragment of the plasmid pTrc99A ( 22 ) following production of blunt-ends using Klenow fragment of E.coli DNA polymerase I and dNTPs. The ligated DNA was used to transform E.coli C122 and transformants screened for production of HsdR by means of SDS-PAGE. DNA sequencing was as previously described ( 23 ).

Polymerase chain reaction

The two primers used for PCR amplification of the first 193 bp of the hsdR gene were as below (the unique restriction sites are shown in bold): Primer 1 5'-CTGATGAGT CATATG CGC-3' ( Nde I) Primer 2 5'-GAACAT TGGCCA GCATCG-3' ( Bal I)

Amplification was carried out as previously described ( 24 ), and the product was purified using Promega Wizardtm PCR Preps and cleaved with Nde I + Bal I.

Protein purification

Purification of M [middot] Eco R124I methylase has been described ( 4 ). For purification of the HsdR subunit an overnight (not more than 8-9 h) culture of E.coli C122[pVZR2] was diluted 1:100 in fresh 2* YT media in a 3 l conical flask containing 500 ml of broth (cells grown in a 5 l fermentor were found to produce a substantially lower yield of HsdR protein), and grown to an optical density at 590 nm of 0.4 (~2 h). This culture was then induced by addition of IPTG to a final concentration of 1 mM. The optimal growth period for maximum yield of protein was found to be 3-3.5 h (OD 590nm = 1.5).

Escherichia coli C122[pVZR2] cell pellet (5 g wet weight) was resuspended in 20 ml of 50 mM Tris-HCl pH 8.0; 7 mM [beta]-mercaptoethanol; 1 mM benzamidine; 100 [mu]M PMSF and subjected to ultrasonic disruption on ice using 15 * 25 s bursts, and the cell debris removed by centrifugation to give 350-400 mg of soluble protein. This solution was made to 0.2 M with NaCl, and 10% PEI600 added to a final concentration of 0.4%. Following centrifugation at 15 000 r.p.m. for 15 min in a Beckman JA-21 rotor, 160-200 mg of protein was present in the supernatant. This protein was precipitated with ammonium sulphate (70% saturation) and the pellet collected by centrifugation at 15 000 r.p.m. for 10 min in a Beckman JA-21 rotor. The pellet was dissolved in 10 ml of buffer A (20 mM Tris-HCl pH 8.0, 50 mM NaCl, 1 mM EDTA, 7 mM [beta]-mercaptoethanol) and dialysed against 2 l buffer A. The dialysate was applied on to a 40 cm * 2 cm 2 column of DEAE Sepharose FF and eluted with 50-300 mM NaCl to prevent protein binding, but allowing DNA to bind to the column. Fractions identified by SDS-PAGE as containing a protein with the same molecular weight as HsdR were pooled and loaded onto a 6 cm * 2 cm 2 Bio-Rad Affi-Gel Blue column. Bound proteins were eluted with a 50-1500 mM NaCl gradient at a flow rate of 0.3 ml/min. HsdR was found to elute between 300 and 700 mM NaCl. The pooled fractions containing HsdR were subjected to another ammonium sulphate precipitation, and the pellet obtained redissolved in 3-5 ml of buffer A and dialysed against the same buffer. This was followed by a further ion exchange column, equilibrated with buffer A, again using Fast Flow DEAE Sepharose and bound protein was eluted using 0.05-0.3 M NaCl gradient. Further purification was obtained by gel filtration using a S300 column. A 150 ml bed volume column (60 * 1.6 cm) was equilibrated with 20 mM Tris-HCl pH 8.0; 150 mM NaCl; 1 mM EDTA; 7 mM [beta]-mercaptoethanol. A 0.5 ml aliquot of sample was loaded, containing 5-10 mg/ml protein, and eluted at 12 ml/h.

Protein concentration was determined from the UV absorbance at 280 nm using an extinction coefficient derived from the aromatic amino acid composition (HsdR, E 280 = 91 900 M -1 cm -1 ).

Reconstitution of Eco R124I endonuclease

The reconstituted endonuclease was produced by titration of MTase with purified HsdR. A 2-fold excess of HsdR subunit was mixed with MTase and the components of the mixture separated by S300 gel filtration. Reconstituted endonuclease was obtained as a single peak with a K av = 0.14.

ATPase assay

For the quantification of ATPase activity, the concentration of inorganic phosphate (Pi) released by ATP hydrolysis was measured by a colorimetric assay based on a spectrophotometric quantification of a phosphomolybdate-malachite green complex ( 25 , 26 ).

DNA binding

Binding reactions contained 50 mM Tris-HCl pH 7.5, 100 mM NaCl, 10 mM MgCl 2 , 1 mM ATP and increasing amounts of the HsdR-subunit and 32 P-labelled DNA fragment in a reaction volume of 10 [mu]l. The concentration of DNA was 5 nM. The reactants were incubated at room temperature for 15 min and were analysed by electrophoresis on 5% native polyacrylamide gel (30:0.8 acrylamide:bisacrylamide) run in 1* TBE buffer ( 19 ). The gels were pre-run at 100 V for 60 min, samples were applied and electrophoresis was continued at 200 V for 20 min. Following electrophoresis the gels were dried and analysed using a Molecular Dynamics Phosphor-Imager and also by autoradiography at -70oC with intensifying screens.

The amount of DNA bound by protein was calculated and percent of DNA bound divided by bound + free DNA was plotted against the R-subunit concentration. K d is the amount of protein needed to complex 50% of the DNA.

DNA cleavage and methylation

Digestion of DNA was performed at 37oC in a buffer containing 50 mM Tris-HCl pH 8.0, 1 mM DTT, 10 mM MgCl 2 . SAM was used at a concentration of 0.2 mM where appropriate. HsdR nuclease activity was assayed over a 3 h time period. The endonuclease activity was assayed in a volume of 25 [mu]l that contained 0.5 [mu]g of plasmid DNA (unless stated otherwise), and varying amounts of enzymes, as described. Reactions were started by addition of ATP to a final concentration of 2 mM. DNA digests were separated on 0.9% TBE/agarose gels.

DNA methylation by the purified enzyme(s) was measured by the transfer of tritiated methyl groups from [ 3 H] SAM (specific activity 85 Ci/mmol, Amersham) to DNA ( 16 ). This was followed by scintillation counting of labelled DNA bound on DE81 filters as described in Promega `Protocols and Applications Guide'.

RESULTS

Cloning the over-expressed hsdR gene and in vivo function

Initially, two recombinant plasmids (pLP25 and pLP17) were constructed carrying a Bsa I fragment containing the entire hsdR gene (Materials and Methods). Both plasmids were able to complement the MTase produced from pAC3MS to restriction-proficiency, this is the first clear indication that the hsdR gene has its own functional promoter. From these plasmids, construction of the recombinant plasmids pETR124 and pVZR2 were as described in Materials and Methods and Figure 1 . Production of HsdR was monitored by SDS-PAGE following IPTG induction of JM109(DE3)[pETR124]. Table 2 shows that while pETR124 plasmid undoubtedly complements the methylase to restriction-proficiency the level of restriction observed (10 -2 ) was low, and IPTG induction completely destroyed this restriction activity. Consequently the plasmid pVZR2, in which the hsdR gene is expressed from the trc promoter, and which was able to complement MTase to give high levels of restriction of bacteriophage lambda in vivo , was used for subsequent purification of HsdR. DNA sequence analysis of this recombinant plasmid using the PCR primers used in the construction of the plasmid (Fig. 1 ) confirmed that the 5'-region of hsdR was as expected and that we had indeed isolated the complete hsdR gene.

Gel filtration and solution molecular mass

Figure 2 A shows the over-production of HsdR from pVZR2, the over-produced HsdR subunit was found in the soluble fraction. S300 gel filtration of HsdR demonstrated that a single species of molecular weight 120 000 Da was present (Fig. 2 B), which agrees well with the predicted molecular weight of HsdR (119 668 Da). Therefore, the subunit exists in a monomeric form in solution. The position of the protein band, following SDS-PAGE, corresponds to the main HsdR band present in the purified endonuclease ( 26 ).

Table 2 Complementation analysis of HsdR and Mtase produced from pAC3MS
IPTG

e.o.p of bacteriophage [lambda] on bacterial strains

induction

2K

2K[pUNG31]

2K[pAC3MS]

2K[pAC3MS]

2K[pAC3MS]

( Eco R124II)

[pETR124]

[pVZR2]

-

[lambda].0

1.0

10 -4

1.0

10 -2

10 -4

+

[lambda].0

1.0

10 -4

1.0

1.0

N.T.

The efficiency of plating (e.o.p.) of bacteriophage lambda was determined as previously described (3). The powerful T7 g10 promoter, present on pETR124, over-produces the HsdR subunit when induced by IPTG. However, the large amount of HsdR subunit produced appears to prevent the correct function of the endonuclease and no restriction activity is observed in vivo . Even in the absence of IPTG induction complementation by this plasmid was poor.

ATPase activity of HsdR


Figure 2 . ( A ) SDS-PAGE showing over-production of HsdR. Lane 1, protein size marker; lane 2, soluble cell extract of JM109(DE3); lane 3, soluble cell extract of JM109(DE3)[pVZR2]; lane 4, soluble extract of JM109(DE3)[pTrc99A]; lane 5, soluble extract of JM109(DE3)[pETR124]; lane 6, purified HsdR. ( B ) S300 gel filtration purification of HsdR. Molecular mass calibration markers were as described in (4) and K av was calculated from (V e - V o )/(V t - V o ), where V e is the elution volume, V o the void volume and V t the column bed volume.

The HsdR subunit, as a component of a restriction endonuclease, is absolutely required for DNA cleavage and production of the ATPase activity associated with DNA translocation. It is also known that HsdR contains an ATP-binding site. In order to determine if the purified HsdR subunit was capable of independent function it was decided to investigate ATPase activity independently of the presence of the other subunits.

HsdR was incubated with ATP in the presence, or absence, of SAM, Mg 2+ and pDRM-1R DNA. Table 3 shows that the HsdR subunit was indeed capable of ATP hydrolysis. ATPase was also observed when the DNA substrate does not carry a s R124II recognition site (pTZ18R), however, no ATP hydrolysis was observed with an oligoduplex carrying a single s R124I recognition site. DNA specificity of the restriction endonuclease is encoded in the hsdS gene, therefore, the fact that plasmid DNA carrying no s R124II recognition site elicited a positive ATPase activity was not altogether surprising. However, the lack of ATPase activity in the presence of the oligoduplex suggests that the size of the DNA fragment is important for ATPase activity. Table 3 also shows that ATP hydrolysis requires both plasmid DNA and Mg 2+ . The level of ATPase activity observed is substantially lower (30-fold of that shown by the endonuclease (see also Table 4 and ref. 26 ) and is similar to the level of activity observed with type III restriction endonucleases ( 27 ).

DNA binding of HsdR

The requirement for plasmid DNA in the determination of ATPase activity suggested we should investigate the ability of HsdR to interact with DNA. This ability was measured using the technique of gel retardation. The subunit was found not to bind an oligoduplex carrying the s R124I recognition site (Fig. 3 A). This mirrors the observation for ATP hydrolysis, in which no ATPase activity was elicited by an oligonucleotide. Therefore, the experiment was repeated with DNA fragments generated from the plasmid pDRM-1R. The plasmid was cut with Hin fI and the DNA fragments end-labelled using [[gamma]- 32 P]ATP. Increasing concentrations of HsdR subunit were incubated with the plasmid DNA. Fragments of 0.2 and 1.2 kb were seen to be retarded in the gel (Fig. 3 A). That this was due to HsdR binding the DNA was confirmed by extraction of the DNA-HsdR complex from low melting point agarose and, following separation on SDS-PAGE, the protein obtained was the same size as HsdR (data not shown). However, neither of the retarded fragments contain a s R124II recognition site and the specificity shown by the HsdR subunit is not that shown by the intact endonuclease-not surprising considering the weight of evidence showing that DNA specificity lies with the HsdS subunit of the endonuclease. Alignment of the DNA sequence of both fragments using the Wilbur-Lipman, the Lipman-Pearson and the Martinez, Needleman-Wunsch alignments, available in the DNASTAR program MegaAlign, failed to identify any significant (>4 bp) sequence present in both fragments.

Table 3 ATPase activity of the HsdR subunit
Composition of the reaction mixture

Concentration of Pi

Specific activity

(mM)

(mmol Pi/[mu]mol enzyme)

HsdR

0

0

HsdR + Mg 2+

0

0

HsdR + Mg 2+ + SAM

0

0

HsdR + pDRM-1R

0.085

0.155

HsdR + Mg 2+ + pDRM-1R

0.480

0.873

HsdR + Mg 2+ + pDRM-1R + SAM

0.520

0.945

HsdR + Mg 2+ + pTZ18R

0.490

0.891

HsdR + Mg 2+ + 39mer

0

0

Reactions (50 [mu]l) were incubated at 30oC for 25 min. Concentration of HsdR was 0.55 [mu]M. DNAs were present at a concentration of 1.35 [mu]M. ATP and MgCl 2 were 10 mM and SAM was 0.2 mM. The concentration of the reaction product (Pi) was measured using a colorimetric assay (25).


Figure 3 . Gel retardation of DNA fragments from the plasmid pDRM-1R by HsdR. ( A ) Lane 1 shows pDRM-1R plasmid DNA following cleavage by Hin fI and end-labelling with [[gamma]- 32 P]ATP. Lanes 2-8 shows the previous plasmid DNA incubated with 0.05, 0.1, 0.2, 0.3, 0.42, 0.85 and 2.1 [mu]M HsdR, respectively. Lanes 9-11 show the result of incubation of an oligoduplex carrying the s R124I site with 0.05, 0.3 and 2.1 [mu]M HsdR, respectively. ( B ) Graph showing the data obtained following phosphorimager analysis of HsdR bound to the purified 1.2 kb fragment from the above Hin fI digest of pDRM-1R. B represents the bound fraction of radiolabelled DNA and F the free fraction of the same DNA.

The 1.2 kb Hin fI fragment of pDRM-1R (Fig. 3 A) was purified from an agarose gel in microgram quantities, the quantity of DNA present was determined by UV spectrophotometry, and then the fragment was used in a further gel retardation experiment. Figure 3 B shows the retardation of the >1.2 kb fragment with various concentrations of HsdR. Quantification of the gel using phosphorimaging analysis allowed a value of 0.2 [mu]M for the K d to be determined. Therefore, HsdR shows weak binding to DNA, however this binding is not non-specific as only certain fragments were retarded in the gel. Nor is the length of the DNA fragment the sole determinant of this weak specificity as other fragments either shorter or longer were not retarded. The possibility that DNA flexibility is important cannot be ruled out.

The ability of the HsdR subunit to cleave DNA (i.e. to act as a non-specific nuclease) was measured by incubating a range of protein concentrations with cccDNA (using pTZ18R, a plasmid which lacks s R124I or s R124II recognition sites) in the presence and absence of SAM, ATP and Mg 2+ . In all cases the cccDNA was found to slowly disappear in the presence of HsdR leaving a smear of DNA slightly higher in the gel than the nicked circle band, and two discrete small DNA bands (data not shown). This weak nuclease activity was found to be Mg 2+ dependent. In addition, the various recombinant clones originally isolated (including pVZR2) were found to restrict the growth of lambda by between 0.5 and 10 -2 , and this activity was increased following IPTG induction of the trc promoter present on pTrc99A (data not shown). However, this activity was only observed immediately following transformation of the E.coli strain. After continued cultivation this activity was lost, suggesting that the bacteria or the HsdR subunit is changed in order to prevent this potentially deleterious effect.

In vitro complementation and ATPase activity of reconstituted endonuclease

The ATPase and nuclease activity of HsdR indicated that the subunit was functional and encouraged us to attempt reconstitution of the endonuclease in vitro . This was accomplished by mixing a 2-fold excess of HsdR with purified Eco R124I MTase. The mixture was applied to a S300 gel filtration column. HsdR subunit was found to elute with a K av of 0.31, MTase was eluted from the same column with a K av of 0.23 and the reconstituted endonuclease eluted with a K av of 0.14 (data not shown). To demonstrate the presence of a functional endonuclease, the ATPase activity of the reconstituted endonuclease was measured. Table 4 shows high-level ATPase activity in the presence of Mg 2+ , pDRM-1R DNA and SAM. Interestingly, ATPase activity occurs in the absence of SAM and DNA. This result has also been observed with the pure endonuclease produced from bacteria expressing all three hsd genes ( 26 ) where it was also shown that SAM can stimulate the rate of ATPase activity and DNA cleavage.

Figure 4 shows that the reconstituted endonuclease cleaves the plasmid pDRM1-R, and for comparative purposes, wild-type endonuclease ( 26 ) also cleaves this plasmid. In addition, the rates of methylation of pDRM-1R for M [middot] Eco R124I, the purified wild-type endonuclease and the reconstituted endonuclease are equivalent (data not shown). Therefore, we have successfully reconstituted the endonuclease in vitro .

DISCUSSION

The HsdR subunit of a type I restriction endonuclease is absolutely required for restriction activity. This restriction activity is also associated with a powerful ATP hydrolysis during which DNA is translocated by the enzyme-DNA complex prior to DNA cleavage ( 26 , 28 , 29 ). The presence of conserved amino acid sequences within the HsdR subunit, which are also present in putative helicases ( 6 ), suggests that the subunit may be capable of independent enzymatic activity. To test this possibility, we have over-produced HsdR and purified the subunit to homogeneity. Two different promoters were used to over-express the hsdR gene. The T7 g10 promoter of pET3a was found to give over-production but the subunit gave poor complementation with M [middot] Eco R124II methylase in vivo (which was totally destroyed by IPTG induction); while, the P trc promoter of the plasmid pTrc99A gave both high yields of HsdR and good complementation to Res + with MTase. Therefore, the P trc - hsdR combination present on pVZR2 was used to produce the HsdR subunit.

Table 4 . ATPase activity of the reconstituted endonuclease R [middot] Eco R124I
Composition of the reaction mixture

Concentration of Pi

Specific activity

(mM)

(mmol Pi/[mu]mol enzyme)

ENase

0

0

ENase + Mg 2+

0

0

ENase + pDRM-1R*

0.197

0.358

ENase + Mg 2+ + pDRM-1R

1.46

20.86

ENase + Mg 2+ + pDRM-1R + SAM

2.29

32.71

ENase + Mg 2+ + 39mer

0.84

12.00

ENase + Mg 2+ + 39mer + SAM

1.82

26.00

ENase + Mg 2+ + pTZ18R*

1.07

1.95

ENase + Mg 2+ + pTZ18R + SAM*

1.23

2.23

MTase + Mg 2+ + pDRM-1R + SAM

0

0

Reactions (50 [mu]l) were incubated at 30oC for 25 min. The concentration of ENase (or MTase) was 0.07 [mu]M and DNAs were present in a concentration of 0.32 [mu]M (except for reactions marked with an asterisk where the enzyme was 550 nM and DNA was 1.35 [mu]M). ATP and MgCl 2 were 10 mM and SAM was 0.2 mM. The concentration of reaction product (Pi) was measured using a colorimetric assay (25,26).


Figure 4 . Cleavage of pDRM-1R DNA by reconstituted endonuclease. The figure shows the results of agarose gel analysis of cleavage of pDRM-1R (single s R124I site) by either wild-type or reconstituted endonuclease. All reactions were carried out in a volume of 25 [mu]l as described in Materials and Methods. Lane 1, 0.5 [mu]g pDRM-1R DNA; lane 2, 0.5 [mu]g pDRM-1R DNA incubated with 1.3 [mu]g of wild-type ENase; lane 3, 0.5 [mu]g pDRM-1R DNA incubated with 1.3 [mu]g reconstituted ENase.Following over-production, purification of HsdR produced a homogeneous, soluble, monomeric protein of molecular mass 120 000 Da (as measured by S300 gel filtration). Therefore, there is no tendency of the subunit towards oligomerisation. A DNA-dependent, Mg 2+ -dependent ATPase activity of the HsdR subunit was detected. This is the first clear demonstration that this HsdR subunit is capable of independent enzyme activity. This activity undoubtedly reflects the presence of an active Walker type I ATP-binding site in the subunit, and may well involve the `DEAD' boxes ( 6 ).

The HsdR subunit was also found to be able to cleave plasmid DNA and therefore behaves as a weak nuclease. That this activity was not due to a contaminating enzyme was demonstrated following assembly of the HsdR subunit into an active endonuclease in vitro . The endonuclease did not cleave pBR322 (no s R124I or s R124II sites), or fully modified pDRM-1R at enzyme:DNA ratios up to 10, while under similar conditions these plasmids were cleaved by pure HsdR. This nuclease activity is quite weak, and a very slow reaction requiring some 4 h for complete digestion of cccDNA. However, the nuclease activity was also found to occur in vivo , and bacteriophage lambda was restricted by up to two orders of magnitude in the presence of the induced pVZR2 plasmid, immediately following transformation of E.coli . This nuclease activity in vivo may explain difficulties we have previously observed when isolating the hsdR gene (K. Firman, unpublished observations). This may also explain why over-production of HsdR, from the powerful T7 g10 promoter, results in loss of complementation with MTase-the detrimental effect of having a higher concentration of HsdR over MTase may affect growth of the bacterial cell in such a way as to prevent restriction activity. This nuclease activity suggests that the endonuclease may have evolved from the methylase by acquisition of a nuclease subunit.

The HsdR subunit was also screened for DNA binding by means of gel retardation. No retardation of an oligoduplex was observed. However, two of the Hin fI fragments of pDRM-1R were found to be retarded as the concentration of the HsdR subunit was increased. Neither of the retarded fragments carried the s R124I or s R124II recognition sequences, nor did they carry an origin of replication (which we thought might be preferentially bound by the HsdR subunit). There was no other obvious conserved DNA sequence in the retarded fragments suggesting that size rather than DNA specificity may be the determining factor for DNA binding. Although, since other fragments of similar size were not retarded, DNA bending cannot be ruled out as being important. The 1.2 kb Hin fI fragment of pDRM-1R was retarded at the lowest HsdR concentration. Therefore, this fragment was purified in order to estimate the dissociation constant for this binding. The value obtained, 0.2 [mu]M, indicates that the binding is fairly weak. It is tempting to link this DNA-binding activity with the hydrolysis of ATP, since the level of ATPase activity associated with HsdR is similar to that observed for the type III endonuclease Eco P15I, which is thought to translocate DNA prior to DNA cleavage. It is possible that a small DNA loop is required prior to translocation (the endonuclease has to attach to the DNA at a site other than the recognition sequence before it can begin to translocate). Therefore, it is possible that HsdR is looping DNA in a manner which is also common to type III enzymes, perhaps even wrapping the DNA around the protein. The ability of the DNA to bend may influence such an activity and this may explain why there are certain `preferred' DNA fragments subject to gel retardation by HsdR.

While there is no doubt that the HsdR subunit is capable of ATP hydrolysis, the level of this hydrolysis is much lower than that observed with the reconstituted endonuclease. This may also reflect the DNA-dependence of this activity. The endonuclease is capable of binding specifically to DNA at its recognition site, but the HsdR subunit is not capable of such highly specific binding and binds DNA only weakly. The importance of DNA binding in the production of the ATPase activity is supported by the observation that an oligoduplex containing the s R124I recognition site does not allow HsdR to hydrolyse ATP, but plasmid DNA (with or without the s R124I recognition site) does support this hydrolysis.

Mixing purified M [middot] Eco R124I DNA methyltransferase with excess HsdR, followed by S300 gel filtration, resulted in re-constitution of an active restriction endonuclease. Enzyme activity was detected as both a powerful ATPase activity in the presence of Mg 2+ and DNA and by cleavage, or methylation, of a supercoiled DNA substrate carrying a single s R124I recognition site, in the presence of ATP, SAM and Mg 2+ . As with the purified, over-produced endonuclease both ATPase activity and DNA cleavage were SAM-independent ( 26 ). One interesting observation was that ATPase activity of the reconstituted endonuclease was stimulated by both plasmid pDRM-1R DNA, and by a 30 bp oligoduplex. Since the initial ATPase activity of type I endonucleases has been associated with DNA translocation ( 11 ) the activation by a short DNA fragment suggests that translocation may not be an absolute requirement for this ATPase activity. DNA cleavage is also not an absolute requirement as this cannot occur with such a short DNA fragment. Such a prenuclease ATPase activity has previously been postulated for Eco BI ( 28 ), based on observed ATPase activity with linear molecules carrying s B sites near the terminus.

In conclusion, we have purified the HsdR subunit of the restriction endonuclease R [middot] Eco R124II to homogeneity, and shown that it can function as an independent ATPase. HsdR was also found to capable of weak DNA-binding and DNA-cleavage. This nuclease activity was also found to occur in vivo . We have also demonstrated that this subunit can be used to reconstitute an endonuclease from pure DNA methyltransferase and HsdR, and shown full activity. This is the first clear demonstration of an enzymatic activity for the HsdR subunit and suggests that type I restriction endonucleases have evolved from a site-specific methylase following interaction with the non-specific nuclease encoded by HsdR.

ACKNOWLEDGEMENTS

We would like to thank Darren Mernagh for supplying the plasmid pDRM-1R and David Dryden for helpful comments. The work carried out by the group in Pushchino was supported by a Interlaboratory Collaborative Research Grant from the Wellcome Trust. Dr Agnes Abadjieva was employed on a Wellcome Trust project grant and Dr Pavel Janscak was employed on a Wellcome Trust International Travelling Fellowship.

REFERENCES

1 Wilson, G. G. and Murray, N. E. (1991) Annu. Rev. Genet., 25, 585-627.

2 Bickle, T. A. and Krüger, D. H. (1993) Microbiol. Rev., 57, 434-450.

3 Hubacek, J. and Glover, S. W. (1970) J. Mol. Biol., 50, 111-127. MEDLINE Abstract

4 Taylor, I., Patel, J., Firman, K. and Kneale, G. G. (1992) Nucleic Acids Res., 20, 179-186. MEDLINE Abstract

5 Dryden, D. T. F., Cooper, L. P. and Murray, N. E. (1993) J. Biol. Chem., 268, 13228-13236.

6 Murray, N. E., Daniel, A. S., Cowan, G. M. and Sharp, P. M. (1993) Mol. Microbiol., 9, 133-143.

7 Yuan, R. (1981) Annu. Rev. Biochem., 50, 285-315. MEDLINE Abstract

8 Yuan, R., Bickle, T. A., Ebbers, W. and Brack, C. (1975) Nature, 256, 556-560. MEDLINE Abstract

9 Studier, F. W. and Bandyopadhyay, P. K. (1988) Proc. Natl. Acad. Sci. USA, 85, 4677-4681.

10 Endlich, B. and Linn, S. (1985) J. Biol. Chem., 260, 5729-5738. MEDLINE Abstract

11 Yuan, R., Heywood, J. and Meselson, M. (1972) Nature New Biol., 240, 42-43. MEDLINE Abstract

12 Price, C., Shepherd, J. C. W. and Bickle, T. A. (1987) EMBO J., 6, 1493-1498. MEDLINE Abstract

13 Piekarowicz, A. and Goguen, J. D. (1986) Eur. J. Biochem., 154, 295-298. MEDLINE Abstract

14 Tyndall, C., Meister, J. and Bickle, T. A. (1994) J. Mol. Biol., 237, 266-274. MEDLINE Abstract

15 Price, C., Bickle, T. A., Firman, K. and Glover, S. W. (1989) J. Mol. Biol., 205, 115-125. MEDLINE Abstract

16 Abadjieva, A., Patel, J., Webb, M., Zinkevich, V. and Firman, K. (1993) Nucleic Acids Res., 21, 4435-4443. MEDLINE Abstract

17 Meister, J., MacWilliams, M., Hübner, P., Jütte, H., Skrzypek, E., Piekarowicz, A. and Bickle, T. A. (1993) EMBO J., 12, 4585-4591. MEDLINE Abstract

18 Amitsur, M., Morad, I., Chapman-Shimshoni, D. and Kaufmann, G. (1992) EMBO J., 11, 3129-3134. MEDLINE Abstract

19 Maniatis, T., Fritsch, E. F. and Sambrook, J. (1982) Molecular Cloning. A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

20 Chang, A. C. Y. and Cohen, S. N. (1978) J. Bacteriol., 134, 1141-1156.

21 Firman, K., Price, C. and Glover, S. W. (1985) Plasmid, 14, 224-234. MEDLINE Abstract

22 Amann, E., Ochs, B. and Abel, K.-J. (1988) Gene, 69, 301-315. MEDLINE Abstract

23 Williams, S. A., Slatko, B. E., Moran, L. S. and Desimone, S. M. (1986) Biotechniques, 4, 138-147.

24 Abadjieva, A., Patel, J., Zinkevich, V. and Firman, K. (1994) J. Mol. Biol., 241, 35-43. MEDLINE Abstract

25 Chan, K.-M., Delfert, D. and Junger, K. D. (1986) Anal. Biochem., 157, 375-380. MEDLINE Abstract

26 Janscak, P., Abadjieva, A. and Firman, K. (1996) J. Mol. Biol., 257, 977-991. MEDLINE Abstract

27 Meisel, A., Mackeldanz, P., Bickle, T. A., Kruger, D. H. and Schroeder, D. H. (1995) EMBO J., 14, 2958-2966. MEDLINE Abstract

28 Endlich, B. and Linn, S. (1985) J. Biol. Chem., 260, 5720-5728. MEDLINE Abstract

29 Yuan, R., Hamilton, D. L. and Burckhardt, J. (1980) Cell, 20, 237-244. MEDLINE Abstract

30 Yanisch-Perron, C., Vieira, J. and Messing, J. (1985) Gene, 33, 103-119.

31 Jacob, F. and Wollman, E. L. (1954) Ann. Inst. Pasteur, 87, 653-673.

32 Mead, D. A., Szczesna-Skorupa, E. and Kemper, B. (1986) Protein Engng, 1, 67-74.

33 Hughes, S. G. (1977). PhD, University of Edinburgh.

34 Patel, J., Taylor, I., Dutta, C. F., Kneale, G. G. and Firman, K. (1992) Gene, 112, 21-27. MEDLINE Abstract

Return

* To whom correspondence should be addressed. Tel: +44 1705 842059; Fax: +44 1705 842053; Email: firmank@biol.port.ac.uk + Present address: Department of Chemistry, University of Portsmouth, St. Michael's Building, White Swan Road, Portsmouth PO1 2DT, UK
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Microbiol. Mol. Biol. Rev.Home page
J. Youell and K. Firman
EcoR124I: from Plasmid-Encoded Restriction-Modification System to Nanodevice
Microbiol. Mol. Biol. Rev., June 1, 2008; 72(2): 365 - 377.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
M. D. Lavigne, L. Yates, P. Coxhead, and D. C. Gorecki
Nuclear-targeted chimeric vector enhancing nonviral gene transfer into skeletal muscle of Fabry mice in vivo
FASEB J, June 1, 2008; 22(6): 2097 - 2107.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (136K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Zinkevich, V.
Right arrow Articles by Firman, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zinkevich, V.
Right arrow Articles by Firman, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?