Cloning, purification and metal binding of the HNH motif from colicin E7

(1)

Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are encouraged to visit:

http://www.elsevier.com/authorsrights

(2)

Cloning, puriﬁcation and metal binding of the HNH motif from colicin E7

Béla Gyurcsik

^a,^⇑

, Anikó Czene

^a

, Hajnalka Jankovics

^a,1

, Noémi I. Jakab-Simon

^a

, Krystyna S´laska-Kiss

^a,b

, Antal Kiss

^b

, Zoltán Kele

^c

aMTA-SZTE Bioinorganic Chemistry Research Group of Hungarian Academy of Sciences, Department of Inorganic and Analytical Chemistry, University of Szeged, Dóm tér 7, H-6720 Szeged, Hungary

bInstitute of Biochemistry, Biological Research Center of the Hungarian Academy of Sciences, Temesvári krt. 62, H-6726 Szeged, Hungary

cDepartment of Medical Chemistry, University of Szeged, Dóm tér 8, H-6720 Szeged, Hungary

a r t i c l e i n f o

Article history:

Received 20 January 2013 and in revised form 26 March 2013 Available online 4 April 2013

Keywords:

HNH nuclease Ubiquitin fusion Zn²⁺-binding Circular dichroism Mass spectrometry Gel mobility shift

a b s t r a c t

The HNH family of endonucleases is characterized by abbametal-ﬁnger structural motif. Colicin E7 is a representative member of this family containing the strictly conserved HNH motif at its C-terminus.

Structural and biochemical studies suggested that the HNH motif could contain all the residues necessary for metal ion binding and nuclease activity. In this work a 43 amino acid peptide extending from V534 to K576 of colicin E7 and encompassing the HNH motif was cloned and expressed inEscherichia colias a ubiquitin fusion protein. The N-terminal fusion tag was cleaved off by a specific protease, and the HNH peptide was purified free of ubiquitin. Circular dichroism, fluorescence and mass spectrometry showed that the zinc-ion binding affinity of the purified HNH peptide was much weaker than that of the intact nuclease domain suggesting that the N-terminal part of the nuclease domain is essential for stabilizing the structure of the HNH motif. The coordination sphere of the metal ion was found to be not fully equipped by the ligand – leaving a free coordination site for the substrate. Neither DNA binding nor DNAse activity of the purified HNH peptide was detected. Comparison of the glutathion-S-transferase-fused N-terminal deletion mutants of the colicin E7 nuclease domain suggested that the presence of the DNA-binding site is still not sufficient for the catalytic activity.

Introduction

The hydrolytic cleavage of phosphodiester bonds in nucleic acids is a part of a number of biologically important processes.

Nucleases play roles in DNA repair, insertions, deletions, transloca- tions, and in degradation of foreign DNA entering the cell. Nucleas- es are often used in gene engineering research and may find applications in future gene therapy[1]. Association of specific substrate binding and catalysis with distinct domains in certain nucleases inspired the development of artificial nucleases. The most relevant examples of this approach are the chimeric zinc-finger nucleases[2]. In these enzymes zinc-finger arrays recognize specific DNA sequences and the nuclease domain of the FokI restriction endonuclease acts as a catalytic domain. The FokI nuclease domain, by itself, is a non-specific nuclease cleaving double stranded DNA upon dimerization on the substrate[3]. Cytotoxicity displayed by zinc-finger nucleases[4–7]may partly be attributed to the loss of control of the enzymatic action by removing the

DNA binding domain of the native FokI enzyme[8]. This undesir- able property is a major obstacle in their use for gene correction in humans justifying the search for alternative nuclease domains.

Efforts to develop small and efﬁcient new nuclease modules can capitalize on results of studies with metal ion binding synthetic multihistidine peptides displaying nuclease activity [9–11] and on conserved sequences within nuclease families. The HNH family of endonucleases is characterized by a commonbb

a

metal-ﬁnger structural motif. Enzymes belonging to the HNH family of proteins [12–14]play roles in a large number of cellular processes[15–18].

The HNH motif,²first identified in homing endonucleases[19], is the second most common motif (8%) in Type II restriction endonucleases (e.g. MnlI[20]and KpnI[21])[22]. In bacterial toxins (colicins and pyocins)[23,24]the strictly conserved HNH motif (HHX14NX8HX3H) is located close to the C-terminus[12]. Detailed biochemical and structural studies with the nuclease domain of colicin E7 (NColE7) [12,23,25–28]and E9[29–32]including X-ray crystallography, solution NMR, CD spectroscopy and mass spectrometry showed that strong nonspecific DNA binding occurs in the major groove, while the catalytic HNH motif, forming a complex with a metal ion, faces

http://dx.doi.org/10.1016/j.pep.2013.03.015

⇑ Corresponding author.

E-mail address:gyurcsik@chem.u-szeged.hu(B. Gyurcsik).

1 Present address: Bio-Nanosystems Laboratory, Faculty of Information Technology, Research Institute of Chemical and Process Engineering, University of Pannonia, H- 8201 Veszprém, P.O. Box 158, Hungary.

2Abbreviations used: HNH motif, HHX14NX8HX3H; NColE7, nuclease domain of colicin E7; GST, glutathion-S-transferase; MS, mass spectrometry; TOF, time-of-ﬂight.

Contents lists available atSciVerse ScienceDirect

Protein Expression and Puriﬁcation

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / y p r e p

(3)

the minor groove of the DNA. For colicin E7 Zn²⁺[12,23,27,33], while for colicin E9 Ni²⁺[29,31,34–37]was suggested to be the physiolog- ical metal ion. The metal ions – coordinated by the histidine sidechains – function on different levels of the nuclease action, such as substrate binding and recognition, electrostatic activation, generation of the hydroxide ion for nucleophylic attack[38], charge com- pensation and stabilization of the pentavalent intermedier, and protonation of the leaving group. In colicins the fourth, most conserved histidine was suggested to act as the base that generates the hydroxide ion for the nucleophylic attack of the phosphate backbone. There is a debate about the role of the metal ions[27,31].

To study the properties of the HNH motif in the absence of the rest of the nuclease domain, we cloned the 43 amino acid long HNH motif of colicin E7 (V534 through K576, according to the colicin E7 original numbering[25]), and puriﬁed it in high yield from Escherichia colias a chimeric protein fused to ubiquitin. The HNH motif was separated from the fusion partner by a speciﬁc protease cleavage. Zn²⁺-ion and DNA binding as well as nuclease activity of the HNH peptide was studied by a variety of methods. To explore the role of the N-terminal part of the NcolE7 in DNA binding and nuclease activity, N-terminally truncated variants of the nuclease domain were constructed.

Materials and methods

Strains and media

E. coliDH10B FendA1 recA1 galU galK deoR nupG rpsLDlacX74 U80lacZDM15 araD139D(ara,leu)7697 mcrAD(mrr-hsdRMS-mcrBC) k[39]was used as cloning host for recombinant DNA work and E. coliBL21(DE3) FompT gal [dcm] [lon] hsdSB[40]for overexpression of the ubiquitinylated HNH fusion proteins (Ub-HNH and HUE-HNH). Bacteria were grown in LB or (for overexpression of the HNH peptide) in TB medium [41] containing ampicillin (100

l

g/ml) or kanamycin (50

l

g/ml) at 37°C.

Plasmid construction

The DNA segment encoding the HNH motif was amplified by PCR using plasmid pQE70-NColE7/Im7 [12] as template and the oligonucleotides AK197 and AK198 (Table 1) as primers. AK197 contains an NcoI site, whereas AK198 contains a TAG stop codon and a SalI site as 5⁰-extension. The PCR product was cloned into plasmid pTZ57R using InsTAclone PCR Cloning Kit (Fermentas), and its nucleotide sequence was verified by sequencing. The NcoI-SalI fragment was transferred first into pOK-BAD (Kan^R) [42]to yield pOB-HNH, then cloned between the BamHI and SalI sites of pET26b-Ub (Kan^R)[43]to obtain pET26b-Ub-HNH. Ligation of the filled-in BamHI and NcoI ends restored the BamHI and NcoI

sites in pET26b-Ub-HNH and created in-frame fusion between the ubiquitin gene and the gene segment coding for the HNH protein.

To facilitate puriﬁcation of the free HNH peptide, the BamHI – SalI fragment was transferred from pET26b-Ub-HNH into the plasmid vector pHUE[44]. The resulting plasmid pHUE-HNH (Amp^R) encodes an N-terminally His-tagged ubiquitin-HNH fusion protein.

N-terminal deletions in NColE7 were created by PCR-ampliﬁca- tion of segments of the NColE7 gene using combinations of different upstream primers and the same downstream primer (Table 1).

The truncated genes were cloned between the EcoRI and XhoI sites of the glutathion-S-transferase (GST) fusion vector pGEX-6P-1 (GE Healthcare).

The recombinant DNA work used standard methods [41].

Restriction enzymes, Taq polymerase, DNA polymerase Klenow fragment and T4 DNA ligase were purchased from Fermentas or from New England Biolabs.

Protein puriﬁcation

BL21(DE3) cells containing pET26b-Ub-HNH were grown to a density of OD5900.4, then IPTG was added to 1 mM to induce Ub-HNH expression, and growth was continued for 3–5 h. Cells were harvested by centrifugation, then suspended in 30 ml extrac- tion buffer (50 mM potassium phosphate, 100 mM NaCl, 7 mM 2- mercaptoethanol, pH 7.4) and disrupted by sonication. The cell extract was centrifuged at 14,000 rpm for 30 min at 4°C in a Sorvall SS34 rotor, then nucleic acids were precipitated from the supernatant with 1% streptomycin sulphate. After centrifugation, the supernatant was diluted by adding one volume of a buffer containing 50 mM potassium phosphate pH 7.4, 7 mM 2-mercaptoethanol, 5% glycerol, and loaded onto a 9.53.5 cm phosphocellulose (Whatman P11) column equilibrated with the same buffer. Pro- teins were eluted with a 200 ml 0–1 M NaCl gradient. Ub-HNH containing fractions were pooled and subsequently concentrated by ultrafiltration (Amicon Ultracel 10 K). The concentrated Ub- HNH solution was further purified by gel filtration on a Superdex 75 PrepGrade 16/60 (GE Healthcare) column in a buffer containing 50 mM potassium phosphate, 100 mM NaCl, pH 7.4 or in buffers required by further experiments (0.1 M ammonium acetate, pH 8.0 for mass spectrometry, or 10 mM Tris, 100 mM NaCl, pH 7.4 for circular dichroism measurements).

The HUE-HNH fusion protein was prepared using a similar procedure. BL21(DE3) harboring pHUE-HNH was grown and HUE- HNH expression was induced essentially as described above. Cells were resuspended in 20 ml of buffer A (50 mM potassium-phosphate pH 7.4, 300 mM NaCl, 12 mM imidazole, 20 mM 2-mercaptoethanol, 30% glycerol) and disrupted by sonication. After centrifugation and removing the nucleic acids with streptomycin sulphate, the supernatant was applied to a 2.5 ml Ni-agarose (His-Select, Sigma) column equilibrated with buffer A. After wash- ing the column with buffer A containing 20 mM imidazole, HUE- HNH was eluted with 50, 100, 150, 200 and 250 mM imidazole steps. Peak fractions were combined and dialyzed o/n against a buffer containing 50 mM potassium-phosphate pH 7.4, 300 mM NaCl, 2 mM 2-mercaptoethanol and 30% glycerol.

The deubiquitilating enzyme Usp2-cc was prepared from BL21(DE3) cells containing the plasmid pHUsp2-cc as described [44]. To cleave the HNH peptide from its N-terminal fusion partner, the puriﬁed HUE-HNH fusion protein was digested with Usp2-cc and the HNH peptide was separated from the ubiquitin tag and the Usp2-cc enzyme by afﬁnity chromatography on a His-Select column using conditions described in[44].

Fractions obtained during protein puriﬁcation were analyzed by standard SDS gel electrophoresis[41]or – for small proteins – by Tricine–SDS–polyacrylamide gel electrophoresis (Tricine–SDS–

PAGE)[46]using 16% gels and Coomassie staining. Protein concen- Table 1

Primers used in this study. Restriction enzyme sites added as 5⁰-extensions are underlined. For N-terminal deletions of NColE7, the same reverse primer (ColE7 reverse) was combined with different forward primers.

Name Sequence (5⁰?3⁰)

HNH GGCGAATTCTCAGGGAAGAGAACTTCATTCG

4N69-NColE7 GCGAATTCAGGAACAATAATGATCGAATGAAGG 4N45-NColE7 GCGAATTCAGTTTCGATGATTTTCGTAAGAAATTC 4N45-NColE7⁰ GCGAATTCAGTTTCGATGATTTTGGTAAGAAATTC 4N25-NColE7 GGAATTCGACTTAGGTTCTCCTGTTCCA 4N4-NColE7^* GGAATTCCCAGGGAAGGCAACAGGTA

ColE7 reverse GCCGCTCGAGCTATTTACCTCGGTGAATATCAATATGC

AK197 GCCATGGTTTCAGGGAAGAGAACTTCA

AK198 GTCGACTATTTACCTCGGTGAATATCAAT

*The4N4-NColE7 mutant carries a C-terminal modiﬁcation[45].

(4)

tration was estimated with Bradford reagent (Sigma) using bovine serum albumin as standard.

GST-tagged truncated NColE7 variants were purified by affinity chromatography using Glutathione Sepharose 4B resin (GE Health- care) as recommended by the manufacturer. To separate the NColE7 mutants from the GST fusion partner, the purified chimeric proteins were digested with PreScission Protease (GE Healthcare), and GST-free ColE7 nuclease variants were collected as the flow- through fractions in Glutathione Sepharose 4B chromatography.

The concentration of the obtained protein solutions varied between 0.4 and 2.0 mg/ml.

Mass spectrometric measurements

Mass spectrometric (MS) measurements were performed on a Q-TOF Premier instrument equipped with a Z-spray electrospray ion source and a time-of-ﬂight (TOF) analyzer (all from Waters).

The instrument was operated in positive ion mode and it was cal- ibrated using the product ions generated from fragmentation of the doubly charged molecular ion of GFP (m/z785.8). Argon was used as the collision gas. Samples were introduced with a syringe pump of the mass spectrometer using a 250

l

L glass syringe coupled with fused silica capillary tubing. The electrospray needle was adjusted to 3.3 kV and N2was used as a nebulizer gas. A stock solution of 100

l

M Zn(acetate)2 (Sigma) was used to titrate the 20

l

M Ub-HNH or HNH protein solution. The recordedm/zdata were deconvoluted using the MassLynx™ v4.1 (Waters) software equipped with the MaxEnt1 algorithm.

Circular dichroism spectroscopy

Circular dichroism spectra were initially recorded using a Jo- bin–Yvon CD6 dichrograph. The Synchrotron Radiation Circular Dichroism (SRCD) spectra of HNH, Ub-HNH and their metal complexes were recorded at the SRCD facility at the UV1 beamline of the storage ring ASTRID at the Institute for Storage Ring Facilities, University of Aarhus, Denmark[47]. Camphor-sulfonic acid served as a calibration material for the instrument. All spectra were recorded with 1 nm steps and a dwell time of 2 s per step, using 0.1 mm quartz cells (SUPRASIL, Hellma GmbH, Germany), for the wavelength range of 175–260 nm. The substances were dissolved in Tris–HCl buffer, pH 7.4, and the pH was adjusted by HCl and NaOH solutions where needed. The protein concentrations were 1.010⁵M. From raw spectra the water baseline was subtracted.

Fuorimetric measurements

Fluorimetric measurements were performed on a Hitachi F- 4500 instrument in aqueous solutions, using a 11 cm quartz cell.

During the measurements the fluorescence of the zinc(II) – TFLZn complex was followed. The ligand TFLZn (Sigma) emits light of 510 nm upon excitation at 360 nm when it forms complex with zinc(II)[48], whereas the ligand is not fluorescent under these conditions. The fluorescence decreases upon addition of a competing molecule, which can bind zinc(II) more strongly than the TFLZn.

The concentraction of TFLZn was, in most experiments 1.010⁵M, and the concentrations of zinc(II), the HNH protein, plasmid DNA and EDTA were varied.

DNase activity assay

DNase activity of puriﬁed Ub-HNH and HNH was tested by incu- bating0.1–0.2

l

g supercoiled pUC18 plasmid DNA with 0.1

l

g of puriﬁed Ub-HNH or 1–3

l

g HNH peptide in 10 mM Tris–HCl pH 9.0 with or without Zn²⁺-ions at 50°C for 20 min. ZnCl2concentrations varied between 0.05 and 120

l

M. Products of the reactions were

analyzed by electrophoresis in 17 cm long 0.7% agarose gels in Tris–borate buffer[41]. For good separation of plasmid forms, the gels were run overnight at 25 V, and stained with ethidium bromide after the run.

Gel electrophoretic mobility shift assay

Gel electrophoretic mobility shift assays were performed with puriﬁed Ub-HNH, free HNH peptide, GST-tagged and free NColE7 variants.

For Ub-HNH and HNH, Csp6I-generated fragments of pUC18 (1739, 676 and 271 bp) were³²P-labeled by polynucleotide kinase.

The labeled fragments were incubated with 0.025–7.5

l

g Ub-HNH or 0.3–6.0

l

g HNH peptide in Tris–HCl pH 7.5 buffer with or without ZnCl₂at room temperature for 15 min, then were separated by electrophoresis in 6% polyacrylamide gels in TBE buffer. The concentration of Zn²⁺ions varied between 1 and 12

l

M.

For truncated NColE7 variants, a 4658 bp plasmid containing tandem repeats of a 196 bp insert (from Jesper Svejstrup, Cancer Research, UK) was used as DNA substrate for gel mobility shift assays. EcoRI digestion of this plasmid generates a2700 bp fragment (plasmid backbone) and multiple copies of the insert fragment. Binding reactions contained 50–100 ng of EcoRI-digested plasmid DNA, puriﬁed GST-NColE7 or free NColE7 variants in 50 mM Tris–HCl, pH 7.4 or PBS (50 mM NaCl, 27 mM KCl, 100 mM Na2HPO4, 18 mM KH2PO4, pH 7.3) and 5% glycerol. The NColE7: DNA binding site molar ratio varied between 0.5:1 and 50:1 whereas the protein: Zn²⁺molar ratio was 1:1. A10 bp long sequence covered by NColE7 upon binding to DNA was considered as a non-speciﬁc DNA binding site[23,27,28]. The reaction mix- tures were incubated for 20–40 min at room temperature, then loaded onto 1% agarose gels and run at 30–50 V in TBE buffer (44.5 mM Tris, 1 mM EDTA, 44.5 mM boric acid, pH 7.3). The gel was stained with ethidium bromide after the run.

Results

Cloning and puriﬁcation of the HNH motif

Previous structural and biochemical studies charaterized the roles of the HNH motif in metal ion binding, DNA binding and nuclease activity of colicin E7[23,26,33,49]. To test these functions with the HNH motif separated from the rest of the nuclease domain, we cloned the C-terminal segment of the ColE7 gene that encodes the 43 amino acid peptide V534SGKRTSFELHHEKPISQNG GVYDMDNISVVTPKRHIDIHRGK₅₇₆ encompassing the HNH motif.

The conserved amino acids – histidines responsible for metal ion binding and initiating the nucleophylic attack and the asparagine stabilizing the structure of the HNH motif – are underlined. The DNA segment encoding the 43 amino acid peptide was synthesized by PCR and cloned in the plasmid vector pOK-BAD to yield pOB- HNH. The pOK-BAD vector was selected for cloning because it allows tight transcriptional control of the target gene[42,50], which seemed to be important because of the potentially toxic phenotype of the HNH peptide. It is known that the nuclease domain of ColE7 (NColE7) can only be expressed in the presence of its inhibitor, the ColE7 immunity protein Im7[51].E. colicells carrying the pOB- HNH plasmid did not show decreased growth rate even after induction of HNH expression by arabinose, suggesting that either the HNH motif, by itself, without the rest of the ColE7 nuclease domain, does not have nuclease activity, or it was degraded in the cell. Electrophoresis of extracts of arabinose-induced cells in Tri- cine–SDS gels did not show overproduction of the expected HNH peptide. Therefore, we stabilized the peptide by genetically fusing it to ubiquitin. A plasmid (pET26b-Ub-HNH) encoding the Ub-HNH

(5)

fusion protein was constructed by transferring the DNA encoding the 43 amino acid peptide into the expression vector pET26b-Ub.

Cells harboring pET26b-Ub-HNH produced large amounts of solu- ble Ub-HNH upon IPTG-induction (Fig. 1a).

Ub-HNH was purified by a three-step procedure as described in Materials and methods section yielding 25 mg pure Ub-HNH from one liter of culture. In the final purification step the gel filtration chromatogram of Ub-HNH and the Tricine–SDS–PAGE of the peak fractions (Fig. 1b and c) showed only one protein band with an electrophoretic mobility corresponding to the expected molecular mass of Ub-HNH (13641.5 Da). During the initial testing of nuclease activity, Zn²⁺and DNA binding of the Ub-HNH fusion protein it became clear that the presence of the ubiquitin domain may obscure results.

To be able to study the properties of the native (non-fused) HNH peptide, it was necessary to prepare it in ubiquitin-free form. A new plasmid (pHUE-HNH) expressing an N-terminally His-tagged ubiquitin-HNH fusion variant was constructed using the pHUE/

pHUsp2-cc system, which allows cleavage of the hybrid protein and easy removal of the ubiquitin tag as well as the His-tagged deub- iquitylating enzyme Usp2-cc by Ni-agarose chromatography[44].

The HUE-HNH fusion protein was produced inE. colicells, puriﬁed, digested with Usp2-cc, and ubiquitin-free HNH peptide was prepared as described inMaterials and methods section. The yield of puriﬁed HNH peptide was 2.5 mg per liter of culture (Fig. 1d).

Lack of nuclease activity and DNA binding by the HNH peptide in vitro Nuclease activity of puriﬁed Ub-HNH and free HNH obtained from HUE-HNH was tested on supercoiled pUC18 plasmid DNA

as described in Materials and methods section. Digestion conditions were guided by results of a previous study that characterized nuclease activity of NColE7[33]. No nuclease activity was detected (Fig. S2). These results were consistent with observations that expression of the HNH peptide alone or as a chimeric Ub-HNH fusion protein did not affect growth rate of the host bacterium (see above), and suggested that the HNH peptide, by itself, is not sufﬁ- cient for nuclease activity. No evidence for DNA binding of Ub-HNH and HNH was detected by mobility shift assay using ³²P-labeled DNA fragments and polyacrylamide gel electrophoresis (data not shown).

Zinc(II) ion binding of the puriﬁed proteins

It is known that the metal ion binding is essential for the nuclease activity of NColE7[27,33]. Therefore, we also tested this property of the puriﬁed Ub-HNH and HNH proteins by mass and CD spectrometries. The Ub-HNH protein solution incubated with a fourfold amounts of zinc(II) ions for elongated period of time in 0.1 M ammonium acetate buffer (pH = 8.0) showed a characteristic pattern of differently charged ions in ESI MS spectra, i.e. different protonation states with a maximum around the charge of +10.

A selected region of this spectrum (Fig. 2a) displayed two in- tense peaks withm/zvalues of 1365.11 and 1371.48. The calculated molar masses from the above values assuming a charge of +10 are 13641.1 and 13706.8 Da, respectively. The difference between the two values approximates the atomic weight of a coordinated zinc(II) ion. According to Table 2. these peaks can be assigned to the metal ion free Ub-HNH + 10H⁺, and to the complex Ub-HNH + 8H⁺+ Zn²⁺. The usual pH range of the deprotonation

14.4 1 2 3 M

kDa

16.9

6.5 26.6

0 20 40 60 80 mAU

40 60 80 100 120 ml

10 11 12 13 M kDa

14.4 16.9

6.5 26.6

imidazole (mM) 50 100 150 200 250

kDa

16.9 14.4

6.5 3.5 HUE-HNH

HUE HNH

1 2 3 4 5 6 7 M

26.6

(a) (c)

(b) (d)

Fig. 1.(a) Production of Ub-HNH inE. coliBL21(DE3) cells containing pET26b-Ub-HNH plasmid as monitored by Tricine–SDS–polyacrylamide gel electrophoresis of cell extracts. Lane 1, non-induced; lane 2, induced with IPTG, total extract; lane 3, induced with IPTG, supernatant; M, Low-Molecular Weight Marker (Bio-Rad). The calculated molecular weight of Ub-HNH is 13641.5 Da. (b) Gel filtration chromatogram of Ub-HNH on a Superdex 75 column. The curves indicate absorption at 280 (blue line) and 236 nm (red line). (c) Analysis of fractions 10 through 13 (marked by the black bar in panel b) by Tricine–SDS–polyacrylamide gel electrophoresis. M, Low-Molecular Weight Marker (Bio-Rad). All lanes are from the same gel as shown inFig. S1. (d)Cleavage of the HUE-HNH fusion protein by the deubiquitilating enzyme Usp2-cc and purification of the HNH peptide by Ni-agarose affinity chromatography. Analysis of the cleavage products by Tricine–SDS–polyacrylamide gel electrophoresis. Lane 1, HUE-HNH non- digested; lane 2, HUE-HNH digested with Usp2-cc (due to poor staining the lower band corresponding to the HNH peptide is very weak); lanes 3 through 7, fractions eluted from a Ni-agarose column with an imidazole step gradient as indicated; M, Low-Molecular Weight Marker (Bio-Rad). The HNH peptide recovered by the Usp2-cc cleavage carries an N-terminal Ser-Met dipeptide as a result of the cloning procedure and has a calculated molecular mass of 5102.8 Da. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(6)

processes of the histidine sidechains is between pH5.3 and7.5 corresponding to the logk6.3–6.5 (kbeing the protonation micro- constant measured by NMR method [52]). Thus, these donor groups – presumably the zinc(II) coordination sites – are mostly deprotonated at pH 8.0. Therefore, the zinc(II) complex is assumed to be formed without any replacement of protons under these conditions. The different complex formation processes are deﬁned by the following general equation:

Zn^2þþLH^aþ_x þZnLH^ðaþ2bÞþ_xb þbH^þ

wherea= 7–16, andb= 0–3 inTable 2(considering that L is the neutral form of the protein molecule, the relationshipx=a also holds). E.g. the processes with (11,3), (10,2), (9,1) and (8,0) (a,b) pairs all result in the same product (with the same charge) corresponding to the same peak ofm/z= 1371.48. It has to be noted, that the complex formation is not quantitative, since the peak of the free

30.00000000 0.0

m/z

1360 1365 1370 1375 1380 1385 1390 1395 1400 1405

%

0 100

080311_KZPROTZNC 198 (3.668) Sm (Mn, 2x3.00); Cm (115:222) TOF MS ES+

1.03e3 1371.4261

1365.0983

1362.3615 1368.6559

1377.6183

1375.1151

1373.4476 1379.5758 1383.9066

1385.4979 1387.6398

1396.4847

Ub-HNH+10H⁺ 1365.11

% 100

50

0

1360 1380 1400 m/z Ub-HNH+8H⁺+Zn²⁺

1371.48

Ub-HNH+6H⁺+2Zn²⁺

1377.81

30.00000000 0.0

m/z

1360 1365 1370 1375 1380 1385 1390 1395 1400 1405

%

0 100

080311_KZPROTZNC 198 (3.668) Sm (Mn, 2x3.00); Cm (115:222) TOF MS ES+

1.03e3 1371.4261

1365.0983

1362.3615 1368.6559

1377.6183

1375.1151

1373.4476 1379.5758 1383.9066

1385.4979 1387.6398

1396.4847

Ub-HNH+10H⁺ 1365.11

% 100

50

0

1360 1380 1400 m/z Ub-HNH+8H⁺+Zn²⁺

1371.48

Ub-HNH+6H⁺+2Zn²⁺

1377.81

30.00000000 0.0

m/z

850 900 950 10001050 1100115012001250 1300 13501400 1450 1500155016001650

%

0 100

080311_KZPROTZNC 198 (3.668) Cm (110:232) TOF MS ES+

1.24e3 1371.4808

1371.3033 1371.2076

1246.9164 1246.7211

1246.6691 1143.0026 1142.9154 1055.1461

1142.8529 1246.5648 1143.2643

1143.3268 1246.9945

1371.1122 1247.08571365.1119 1365.0165 1247.1768

1252.6252 1252.8994 1253.0039

1371.6036 1371.6993 1377.8099

1378.0016

1530.5236 1378.1794

1530.4513 1381.5074

1384.2770 1523.4354 1384.4142

1537.4542 1537.7579 1539.5510 1548.3152

1000 1500 m/z

% 100

50

0

pH = 3.5 pH = 8.0

835 840 845 850 855 860 865 870 875 880

851.4481 851.2760

851.1146

850.9534 851.6096

851.7817

851.9540

852.1154

861.9485 861.6129 854.2800

862.2733 862.4465 862.5981 862.7822 867.9449 a

a

a a

a a a

a a

a

851.45

861.95

840 850 860 870 880 m/z 0

100

50

%

30.00000000 06-Feb-2009 08:51:23

ink benne

m/z 46 848 850 852 854 856 858 860 862 864 866 868 870 872 874 876 878 880 882 884 886

HNHZN 41 (0.828) Cm (37:103) TOF MS ES+

2.90e4 851.4266

851.2653

851.1039

850.9318 851.5988

851.7710

861.9269

851.9431 861.5913

861.4181 852.1154

857.6024 852.2768

857.4296 854.1077

854.4309 857.2570

857.9372 858.0992 858.2612 858.4340 862.0892

862.2625

862.4248

862.5981

872.4152 862.7604

872.0775 868.2491 862.9229 871.9142

868.7598 872.7418

872.9052 873.0795 873.2538

878.7429

850 860 870 880 m/z 851.43

861.93

872.42

30.00000000 06-Feb-2009 08:51:23

ink benne

m/z 46 848 850 852 854 856 858 860 862 864 866 868 870 872 874 876 878 880 882 884 886

HNHZN 41 (0.828) Cm (37:103) TOF MS ES+

2.90e4 851.4266

851.2653

851.1039

850.9318 851.5988

851.7710

861.9269

851.9431 861.5913

861.4181 852.1154

857.6024 852.2768

857.4296 854.1077

854.4309 857.2570

857.9372 858.0992 858.2612 858.4340 862.0892

862.2625

862.4248

862.5981

872.4152 862.7604

872.0775 868.2491 862.9229 871.9142

868.7598 872.7418

872.9052 873.0795 873.2538

878.7429

850 860 870 880 m/z 851.43

861.93

872.42

(a)

(b)

Fig. 2.(a) A selected region of the ESI MS spectrum of the Ub-HNH protein solution incubated with a fourfold amount of zinc(II) ions at 4°C for 90 h in 0.1 M ammonium acetate buffer (pH = 8.0). The insert shows the ESI MS spectrum in the 850–1650 m/z range with the selected region indicated by the dashed ellipse. (b) ESI MS spectra of the HNH peptide incubated with fourfold amounts of zinc(II) ions at room temperature for 2 h. Left panel: at pH = 3.5. Right panel: in 0.1 M ammonium acetate buffer (pH = 8.0).

Table 2

The calculatedm/zvalues for Ub-HNH in different protonated states in the absence of metal ion, and in the presence of it assuming different number of protons released upon the zinc(II) binding to three histidine side-chains.

Initial no. of protons 16 15 14 13 12 11 10 9 8 7 6

Ub-HNH 853.59 910.43 975.39 1050.35 1137.79 1241.14 1365.15 1516.72 1706.19 1949.79 2274.58

+Zn^2+a 914.57 979.82 1055.12 1142.96 1246.77 1371.35 1523.61 1713.94 1958.64 2284.92 2741.70

+Zn^2+b 857.47 914.57 979.82 1055.12 1142.96 1246.77 1371.35 1523.61 1713.94 1958.64 2284.92

+Zn^2+c 807.09 857.47 914.57 979.82 1055.12 1142.96 1246.77 1371.35 1523.61 1713.94 1958.64

+Zn^2+d 762.31 807.09 857.47 914.57 979.82 1055.12 1142.96 1246.77 1371.35 1523.61 1713.94

+2Zn^2+d 689.28 725.50 765.75 810.74 861.34 918.70 984.25 1059.88 1148.13 1252.41 1377.55

a Assuming that the metal ion replaces three.

b Two.

c One protons.

d Assuming that the metal ions do not replace protons.

(7)

Ub-HNH protein can also be detected. This is even more expressed when smaller zinc(II) excess is applied (Fig. S3). On the other hand, inFig. 2a the peak of the Ub-HNH containing two zinc(II) ions can also be identiﬁed, but its intensity is signiﬁcantly smaller than that of the peak of the mono complexes. Even at 20-fold metal ion excess only traces of this latter species were observed.

Similar results were obtained in solutions containing the HNH motif and zinc(II) ions. In the ESI MS spectra recorded at pH = 3.5 and 8.0 the signals of both the zinc(II) bound protein and the metal ion-free HNH were observed. The ratio of the two signal intensities increased signiﬁcantly with the increase of pH (Fig. 2b). Further- more, at pH = 8.0 the adduct with two zinc(II) ions was detected in minor amounts. MALDI-MS detected a zinc(II) adduct with 1:1 ratio, but it shall be noted that in this case the matrix itself is able to bind zinc(II). Even at 20-fold excess of metal ion the major peak was assigned to the free HNH, and according to the minor peak only one metal ion was bound to the protein (data not shown).

Circular dichroism spectra recorded on a commercial instrument showed no signiﬁcant difference between the CD patterns of Ub-HNH and Zn²⁺ – Ub-HNH solutions at pH = 7.4. However, the repeated measurements at a synchrotron radiation CD (SRCD) facility - that provided more precise data at wavelengths below 200 nm – showed small but reproducible difference between the spectra of the Ub-HNH protein in the presence and in the absence of zinc(II) ions. The addition of zinc(II) ions to the protein solution resulted in a positive shift of the band intensity in the 190–200 nm wavelength range (Fig. 3a). This change was reversible, as the addition of EDTA excess resulted in the same spectrum as that of the free Ub-HNH. A similar small effect was observed for the colicin E9 nuclease domain upon zinc(II) binding [35] indicating a zinc(II)-induced change in the secondary structure.

The relatively small change in the CD signal for both Ub-HNH and colicin E9 reﬂects the relatively large contribution to the CD

spectra of the rest of the proteins (the ubiquitin in the present case or the N-terminal part of colicin E9), as the organized structures of these domains are not affected by metal ion binding.

This was also approved by the similar character of positive shift of the CD intensity at low wavelengths (k= 180–200) in the spectra of the HNH solution upon addition of two equivalents of zinc(II) (Fig. 3b) This suggested that the small change in the spectrum of the Ub-HNH is related to the change of the secondary structure in the HNH part.

The calculated fractions of the secondary structure elements using the CDPro program package[53] (Table 3.) showed an increase in the fraction of the organized elements, such as the

a

-helix and b-strands upon zinc(II)ion coordination in aqueous solution, and in the presence of 20% TFE. However, in 80% TFE, where the free HNH protein preferentially forms

a

-helical structure, the decrease of this secondary element can be observed upon addition of the metal ion. These observations suggest that independently of the starting conformation, zinc(II) binding induces the peptide to fold into a structure similar to the one detected by crystallography[12,23,25,27].

As no change was observed in the SRCD pattern upon addition of zinc(II)-ions at pH = 3.5, we can conclude, that under these conditions no metal ion binding occurred.

Fluorescence spectroscopic measurements were performed to check, whether the HNH protein is able to compete with the TFLZn fluorescent dye for zinc(II) ions. In this case it is expected, that upon addition of HNH the zinc(II) ions will be removed from the Zn-TFLZn complex (similarly to EDTA – data not shown) and therefore, the fluorescence of the Zn-TFLZn complex will disappear. In- stead we have only observed a slight decrease of the fluorescence at 1:2:2 Zn:TFLZn:HNH molar ratio (Fig. S4). The formation of var- ious complexes, among them ternary species indicated that the protein was not able to fully occupy the coordination sites of the metal ion: out of the number of available coordinating donor groups at most three histidines could be coordinated similarly to the native enzyme.

Functional mapping of the NColE7 N-terminal part

The results described above showed, that the HNH peptide alone, without the rest of the nuclease domain does not have nuclease activity. Moreover, we could not detect DNA-binding by the isolated HNH domain and Zn²⁺-binding capacity of the HNH peptide was found to be weaker than that of the intact NColE7.

These observations suggested that the N-terminal part of the nuclease domain is essential for NColE7 function. To investigate its role in ColE7 function, N-terminal deletion mutants (4N69, 4N45,4N25 and4N4) of NColE7 lacking the indicated numbers of amino acids were constructed, and cloned in the pGEX-6P-1 plasmid vector as described in Materials and methods section.

Two of these fusion plasmids (4N4 and4N25] were constructed in an earlier study[45]. The recombinant plasmids were designed to encode fusions between the glutathione-S-transferase tag (GST-

(a)

-4.0 -2.0 0.0 2.0

180 200 220 240 260

λ/nm

CD/mdeg

Ub-HNH, pH = 7.4 5 eqs. Zn²⁺, pH = 7.4 5 eqs. EDTA, pH = 7.4

(b)

-4.0 -2.0 0.0

180 200 220 240 260

λ/nm

CD/mdeg

HNH, pH = 7.5 HNH, pH = 3.5 2eqs. Zn²⁺, pH = 3.5 2eqs. Zn²⁺, pH = 7.5

Fig. 3. (a) SRCD spectra of the Ub-HNH protein in the absence (red) and presence (pink) of ﬁve equivalents of zinc(II)ions, and upon the addition of EDTA excess to the zinc(II) containing system (yellow). (b) SRCD spectra of the HNH peptide. The effect of the pH on the protein conformation in the absence and presence of zinc(II) ions is demonstrated at pH = 3.5 and 7.5. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

Table 3

The calculated fractions (%) of the secondary elements by the CDPro program package [53]. The spectra of the HNH and Zn-HNH systems recorded in the buffer solution (2 mM Tris, 20 mM NaCl, pH 7.4) and the same buffer containing 20% or 80% triﬂuoro- ethanol (TFE).

Buffer 20% TFE 80% TFE

HNH Zn-HNH HNH Zn-HNH HNH Zn-HNH

a-helix 6.0 7.0 16.0 21.0 40.0 34.0

b-strand 16.0 28.0 20.0 23.0 11.0 17.0

Turn 23.0 24.0 23.0 22.0 22.0 20.0

Other 55.0 41.0 41.0 34.0 27.0 29.0

(8)

tag) and the N-termini of the truncated NColE7 variants. A plasmid encoding a GST-fusion with the HNH peptide was also constructed for comparison. Members of the deletion series contained different functional elements of NColE7. The4N69 mutant contained, besides the HNH motif, the Im7 binding site, whereas the 4N45 and4N25 mutants also contained the DNA binding helices. The 4N4 mutant contained all elements mentioned above and lacked only the N-terminal, positively charged tetrapeptide KRNK[45].

Surprisingly, expression of all four NColE7 deletion derivatives was successful inE.coli indicating, that the mutant enzymes did not have cytotoxic activity (Fig. S5). Under the same conditions the pGEX-6P-1 plasmid containing the gene of the native NColE7 was cytotoxic forE. coli.

The fusion proteins were purified by affinity chromatography and DNA electrophoretic mobility shift assays were performed as described inMaterials and methods section. Among the chimeric proteins only GST-4N4-NColE7-Ccaused significant decrease of mobility of the DNA (Fig. 4). This was evident with the GST-fused protein, as well as with the free4N4-NColE7-Cnuclease domain, separated from the GST affinity tag by specific protease digestion and subsequent affinity chromatography. Probably due to the size difference between the two proteins, the observed mobility shift was smaller with the free4N4-NColE7-Cnuclease domain than with the fusion protein (Fig. 4). The GST protein itself did not affect the gel mobility of the dsDNA. The smeared appearance of the bands indicated the heterogeneity of the complexes resulting from partial occupation of multiple non-specific binding sites on the fragments. The presence or absence of Zn²⁺-ions did not show significant effect on the gel mobility shift (Fig. S6).

Discussion

The HNH peptide is present in different endonucleases and comprises a well conserved bb

a

metal-ﬁnger structural motif forming the active center[17]. This motif is located at the C-terminus of bacterial toxins, such as the Colicin E7 nuclease. Results of previous studies could be interpreted to mean that the HNH motif contained all or most of the structural elements responsible for metal and DNA binding and DNAse activity of the ColE7 nuclease.

Here we tested this assumption by cloning and puriﬁcation of the HNH peptide, and testing its metal and DNA binding capacity and nuclease activity. Initial attempts to purify the HNH peptide failed because of its degradation inE. coli. We could stabilize the HNH peptide by genetically fusing it to ubiquitin. The Ub-HNH fusion protein could be puriﬁed in high yield adding a new example to

the list of successful applications[44,54]of the ubiquitin tag for stabilizing foreign proteins inE. coli. To be able to study the un- tagged HNH peptide, the ubiquitin tag was split off by specific protease cleavage and the HNH peptide was purified. It has to be noted that if the protease cleavage was performed in the bacterial cells, the ESI MS spectra of the raw cell extracts (Fig. S7) showed only peaks characteristic for the different charge states of the ubiquitin (Mr= 8556.7 Da). This clearly indicated that the free HNH peptide was degraded inE. colicells. Therefore, we used an expression system[44], which involves cleavage of the purified fusion protein and purification of the HNH peptide free of ubiquitin. This strategy allowed us to purify the HNH peptide in a good yield (2.5 mg from a liter of culture).

Possible reasons for the lack of nuclease activity of the HNH peptide include: (i) weak metal ion binding – (the zinc(II)-binding is essential for nuclease activity); (ii) the improper structure of the catalytic site, (iii) lack of DNA substrate binding.

Metal binding was studied with free HNH peptide as well as with the ubiquitin-fused Ub-HNH protein. Observation of consid- erable amounts of the free protein besides the zinc(II) containing species by ESI MS suggested that both the Ub-HNH protein and the HNH peptide show lower afﬁnity towards zinc(II) ions than the colicin E9 protein, for which the dissociation constant was determined to be less than 1.0 nM[36]. This can probably be ex- plained by the structural pattern of the HNH motif. The ESI MS of the colicin E9 nuclease domain showed a bimodal charge state, and the zinc(II) ions were bound by the protein in the more com- pact conformation[35]. Such phenomenon was not observed with Ub-HNH, showing that the ubiquitin fusion partner did not stabilize the HNH motif.

Zinc(II) ions induced structural changes in the HNH motif were detected by SRCD spectroscopy. The structure obtained in crystals by X-ray crystallography[12,23,25,27]is partly formed in aqueous solution (Table 3.). However, even if the Zn-HNH complex is not formed to 100% under the conditions used, at least a fraction of the HNH peptide in aqueous solutions is likely to be in complex with Zn. Therefore, the lack of the catalytic activity can probably be attributed more to impaired DNA binding than to poor metal binding. This is also supported by the fact that the strong DNA binding region, which overlaps with the Im7 protein (the inhibitor of ColE7) binding site is outside the HNH motif (Table 4.)[49,55].

DNA 20×GST-ΔN69-NColE7 20×ΔN69-NColE7 5×GST-ΔN45-NColE7 5×ΔN45-NColE7 9×GST-ΔN25-NColE7 9×ΔN25-NColE7 28×GST-ΔN4-NColE7-C* 28×ΔN4-NColE7-C* 22×GST-HNH 22×HNH

2700bp

196bp Fig. 4.DNA binding by N-terminally truncated variants of NColE7 before and after digestion with PreScission Protease. Gel-mobility shift assay was performed as described inMaterials and methods sectionusing 1% agarose gel and ethidium bromide staining. The excess of GST-fused or free NColE7 variants used in the different samples is indicated above the lanes.

Table 4

The interacting amino acids of the NColE7 with the DNA substrates and with its immunity protein Im7 according to the crystal structures published. The three amino acid residues in bold play role both in DNA and Im7 binding.

DNA[23,27,28] Im7[49]

5⁰-GCGATCGC- 3⁰

5⁰-CGGGATATCCCG- 3⁰

5⁰-. . .CGATCGAA. . .- 3⁰

K490

D493 D493 S514

D494 R515

R496 R496 R496 N516

K497 N517

K498 D519

R520 R520 R520 R520

K525 K525 K525 K525

A526 A526 A526 K528

T529 R530

S535 T531

G536 G536 Q532

K537 K537 K537 K537

R538 R538 R538 T539

S540 S540 S540

E542 E542

L543 L543 L543

H544 H544

(9)

Although the fluorescence spectroscopic measurements suggested that the HNH peptide does not fill the inner coordination sphere of the metal ion – leaving an empty coordination site for the substrate, this binding is very weak. Indeed, N-terminally truncated mutants of NColE7 caused significant gel mobility shift of the DNA only when the whole DNA binding region was reconstructed.

The ESI MS experiments further showed the presence of trace amounts of binuclear zinc(II) complexes that suggest a second, weak metal ion binding site in the HNH motif. This is in agreement with the observation that addition of excess zinc(II) inhibits the catalytic activity of the ColE7, presumably because of binding to the fourth histidine moiety, and preventing its role as a general base in the catalytic process[33].

It has to be mentioned that among the truncated mutants even the one lacking only four amino acids at the N-terminus was cata- lytically inactive. This suggests that the very N-terminus of the protein is necessary for the activity of the C-terminal catalytic center. Based on this property we suggest that the HNH motif may form the catalytic center of an appropriately constructed novel articficial chimeric nuclease with allosterically controlled catalytic activity. HNH nucleases have recently been applied as promising alternatives of the FokI nuclease domain in artificial nucleases demonstrating that the new chimeric nucleases nick the DNA pre- dominantly site-specifically[56]. Therefore, we plan further experiments to fuse the HNH motif and the truncated NColE7 mutants to specific DNA binding proteins, such as zinc-fingers to investigate their effect on the catalytic activity. In parallel, computer modeling of the possible control mechanism will be performed.

Acknowledgments

We thank Kin-Fu Chak for plasmid pQE70-NColE7/Im7, Rohan T. Baker for plasmids pHUE and pHUsp2-cc, and Ibolya Anton for the technical assistance. This work was supported by the Hungar- ian Scientiﬁc Research Fund (OTKA-NKTH) Grant CK80850, TÁMOP-4.2.1/B-09/1/KONV-2010-0005 and TÁMOP-4.2.2/B-10/1- 2010-0012. FP6 Marie Curie European Reintegration Grant (MERG-CT-2005-022342) for H.J. is greatly acknowledged. The SRCD experiments received funding from the European Commu- nity’s Seventh Framework Programme (FP7/2007-2013) under grant agreement No 226716.

Appendix Supplementary data

Supplementary data associated with this article can be found, in the online version, athttp://dx.doi.org/10.1016/j.pep.2013.03.015.

References

[1] B. Gyurcsik, A. Czene, Towards artiﬁcial metallonucleases for gene therapy:

recent advances and new perspectives, Fut. Med. Chem. 3 (2011) 1935–1966.

[2] Y.-G. Kim, J. Cha, S. Chandrasegaran, Hybrid restriction enzymes: zinc ﬁnger fusions to Fok I cleavage domain, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 1156–

1160.

[3] D.A. Wah, J. Bitinaite, I. Schildkraut, A.K. Aggarwal, Structure of FokI has implications for DNA cleavage, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 10564–

10569.

[4] T.I. Cornu, S. Thibodeau-Beganny, E. Guhl, S. Alwin, M. Eichtinger, J.K. Joung, T.

Cathomen, DNA-binding speciﬁcity is a major determinant of the activity and toxicity of zinc-ﬁnger nucleases, Mol. Ther. 16 (2008) 352–358.

[5] E.-M. Händel, S. Alwin, T. Cathomen, Expanding or restricting the target site repertoire of zinc-ﬁnger nucleases: the inter-domain linker as a major determinant of target site selectivity, Mol. Ther. 17 (2009) 104–111.

[6] T. Cathomen, J.K. Joung, Zinc-ﬁnger nucleases: the next generation emerges, Mol. Ther. 16 (2008) 1200–1207.

[7] S.M. Pruett-Miller, D.W. Reading, S.N. Porter, M.H. Porteus, Attenuation of zinc ﬁnger nuclease toxicity by small-molecule regulation of protein levels, PLoS Genet. 5 (2009) e1000376.

[8] D.A. Wah, J.A. Hirsch, L.F. Dorner, I. Schildkraut, A.K. Aggarwal, Structure of the multimodular endonuclease FokI bound to DNA, Nature 388 (1997) 97–100.

[9] I.N. Jakab, A. Jancsó, T. Gajda, B. Gyurcsik, A. Rockenbauer, Copper(II), nickel(II) and zinc(II) complexes of N-acetyl-His-Pro-His-His-NH2: equilibria, solution structure and enzyme mimicking, J. Inorg. Biochem. 102 (2008) 1438–1448.

[10] I.N. Jakab, O. Lõrincz, A. Jancsó, T. Gajda, B. Gyurcsik, Approaching the minimal metal ion binding peptide for structural and functional metalloenzyme mimicking, Dalton Trans. (2008) 6987–6995.

[11] M.P. Fitzsimons, J.K. Barton, Design of a synthetic nuclease: DNA hydrolysis by a zinc-binding peptide tethered to a rhodium intercalator, J. Am. Chem. Soc.

119 (1997) 3379–3380.

[12] M.J. Sui, L.C. Tsai, K.C. Hsia, L.G. Doudeva, W.Y. Ku, G.W. Han, H.S. Yuan, Metal ions and phosphate binding in the H-N-H motif: crystal structures of the nuclease domain of ColE7/Im7 in complex with a phosphate ion and different divalent metal ions, Protein Sci. 11 (2002) 2947–2957.

[13] H. Huang, H.S. Yuan, The conserved asparagine in the HNH motif serves an important structural role in metal ﬁnger endonucleases, J. Mol. Biol. 368 (2007) 812–821.

[14] M. Punta, P.C. Coggill, R.Y. Eberhardt, J. Mistry, J. Tate, C. Boursnell, N. Pang, K.

Forslund, G. Ceric, J. Clements, A. Heger, L. Holm, E.L. Sonnhammer, S.R. Eddy, A. Bateman, R.D. Finn, The Pfam protein families database, Nucleic Acids Res.

40 (2012) (2012) D290–D301.

[15] E.A. Galburt, B.L. Stoddard, Catalytic mechanisms of restriction and homing endonucleases, Biochemistry 41 (2002) 13851–13860.

[16] P. Mehta, K. Katta, S. Krishnaswamy, HNH family subclassiﬁcation leads to identiﬁcation of commonality in the His-Me endonuclease superfamily, Protein Sci. 13 (2004) 295–300.

[17] J.H. Eastberg, J. Eklund, R. Monnat Jr., B.L. Stoddard, Mutability of an HNH nuclease imidazole general base and exchange of a deprotonation mechanism, Biochemistry 46 (2007) 7215–7225.

[18] A.H. Keeble, M.J. Mate, C. Kleanthous, HNH endonucleases, in: M. Belfort, B.L.

Stoddard, D.W. Wood, V. Derbyshire (Eds.), Homing Endonucleases and Inteins, Springer-Verlag, Berlin, 2005, pp. 49–65.

[19] A.E. Gorbalenya, Self-splicing group I and group II introns encode homologous (putative) DNA endonucleases of a new family, Protein Sci. 3 (1994) 1117–

1120.

[20] E. Kriukiene, J. Lubiene, A. Lagunavicius, A. Lubys, MnlI – the member of H-N-H subtype of Type IIS restriction endonucleases, Biochim. Biophys. Acta 1751 (2005) 194–204.

[21] M. Saravanan, J.M. Bujnicki, I.A. Cymerman, D.N. Rao, V. Nagaraja, Type II restriction endonuclease R.KpnI is a member of the HNH nuclease superfamily, Nucleic Acids Res. 32 (2004) 6129–6135.

[22] J. Orlowski, J.M. Bujnicki, Structural and evolutionary classiﬁcation of Type II restriction enzymes based on theoretical and experimental analyses, Nucleic Acids Res. 36 (2008) 3552–3569.

[23] K.-C.h. Hsia, K.-F. Chak, P.-H. Liang, Y.-S.h. Cheng, W.-Y. Ku, H.S. Yuan, DNA binding and degradation by the HNH protein ColE7, Structure 12 (2004) 205–

214.

[24] Y. Michel-Briand, C. Baysse, The pyocins ofPseudomonas aeruginosa, Biochimie 84 (2002) 499–510.

[25] Y.-S. Cheng, K.-C. Hsia, L.G. Doudeva, K.-F. Chak, H.S. Yuan, The crystal structure of the nuclease domain of colicin E7 suggests a mechanism for binding to double-stranded DNA by the H-N-H endonucleases, J. Mol. Biol. 324 (2002) 227–236.

[26] K.-C.h. Hsia, Ch.-L. Li, H.S. Yuan, Structural and functional insight into sugar- nonspeciﬁc nucleases in host defense, Curr. Opin. Struct. Biol. 15 (2005) 126–

134.

[27] L.G. Doudeva, H. Huang, K.-C.h. Hsia, Zh. Shi, Ch.-L. Li, Y. Shen, Ch.-L. Cheng, H.S. Yuan, Crystal structural analysis and metal-dependent stability and activity studies of the ColE7 endonuclease domain in complex with DNA/Zn²⁺

or inhibitor/Ni²⁺, Protein Sci. 15 (2006) 269–280.

[28] Y.-T. Wang, W.-J. Yang, C.-L. Li, L.G. Doudeva, H.S. Yuan, Structural basis for sequence-dependent DNA cleavage by nonspeciﬁc endonucleases, Nucleic Acids Res. 35 (2007) 584–594.

[29] J.P. Hannan, S.B.-M. Whittaker, A.M. Hemmings, R. James, C. Kleanthous, G.R.

Moore, NMR studies of metal ion binding to the Zn-ﬁnger-like HNH motif of colicin E9, J. Inorg. Biochem. 79 (2000) 365–370.

[30] Kh. Mosbahi, Ch. Lemaitre, A.H. Keeble, H. Mobasheri, B. Morel, R. James, G.R.

Moore, E.J.A. Lea, C. Kleanthous, The cytotoxic domain of colicin E9 is a channel-forming endonuclease, Nat. Struct. Biol. 9 (2002) 476–484.

[31] M.J. Mate, C. Kleanthous, Structure-based analysis of the metal-dependent mechanism of H-N-H endonucleases, J. Biol. Chem. 279 (2004) 34763–

34769.

[32] E.T.J. Van Den Bremer, A.H. Keeble, W. Jiskoot, R.E.J. Spelbrink, C.S. Maier, A.

Van Hoek, A.J.W.G. Visser, R. James, G.R. Moore, C. Kleanthous, A.J.R. Heck, Distinct conformational stability and functional activity of four highly homologous endonuclease colicins, Protein Sci. 13 (2004) 1391–1401.

[33] W.-Y. Ku, Y.-W. Liu, Y.-C. Hsu, C.-C. Liao, P.-H. Liang, H.S. Yuan, K.-F. Chak, The zinc ion in the HNH motif of the endonuclease domain of colicin E7 is not required for DNA binding but is essential for DNA hydrolysis, Nucleic Acids Res. 30 (2002) 1670–1678.

[34] A.H. Keeble, A.M. Hemmings, R. James, G.R. Moore, C. Kleanthous, Multistep binding of transition metals to the H-N-H endonuclease toxin colicin E9, Biochemistry 41 (2002) 10234–10244.

[35] E.T.J. Van Den Bremer, W. Jiskoot, R. James, G.R. Moore, C. Kleanthous, A.J.R.

Heck, C.S. Maier, Probing metal ion binding and conformational properties of the colicin E9 endonuclease by electrospray ionization time-of-ﬂight mass spectrometry, Protein Sci. 11 (2002) 1738–1752.

(10)

[36] A.J. Pommer, U.C. Kuhlmann, A. Cooper, A.M. Hemmings, G.R. Moore, R. James, C. Kleanthous, Homing in on the role of transition metals in the HNH motif of colicin endonucleases, J. Biol. Chem. 274 (1999) 27153–27160.

[37] J.P. Hannan, S.B. Whittaker, S.L. Davy, U.C. Kuhlmann, A.J. Pommer, A.M.

Hemmings, R. James, C. Kleanthous, G.R. Moore, NMR study of Ni²⁺binding to the H-N-H endonuclease domain of colicin E9, Protein Sci. 8 (1999) 1711–

1713.

[38] J.A. Cowan, Metal activation of enzymes in nucleic acid biochemistry, Chem.

Rev. 98 (1998) 1067–1087.

[39] S.G. Grant, J. Jessee, F.R. Bloom, D. Hanahan, Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 4645–4649.

[40] F.W. Studier, A.H. Rosenberg, J.J. Dunn, J.W. Dubendorff, Use of T7 RNA polymerase to direct expression of cloned genes, Methods Enzymol. 185 (1990) 60–89.

[41] J. Sambrook, D.W. Russel, Condensed Protocols from Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2006.

[42] E. Timar, P. Venetianer, A. Kiss, In vivo DNA protection by relaxed-speciﬁcity SinI DNA methyltransferase variants, J. Bacteriol. 190 (2008) 8003–8008.

[43] D.W. Gohara, C.S. Ha, S. Kumar, B. Ghosh, J.J. Arnold, T.J. Wisniewski, C.E.

Cameron, Production of ‘‘authentic’’ poliovirus RNA-dependent RNA polymerase (3D(pol)) by ubiquitin-protease-mediated cleavage inEscherichia coli, Protein Expr. Purif. 17 (1999) 128–138.

[44] A.M. Catanzariti, T.A. Soboleva, D.A. Jans, P.G. Board, R.T. Baker, An efﬁcient system for high-level expression and easy puriﬁcation of authentic recombinant proteins, Protein Sci. 13 (2004) 1331–1339.

[45] A. Czene, E. Németh, I.G. Zóka, N.I. Jakab-Simon, T. Körtvélyesi, K. Nagata, H.E.M. Christensen, B. Gyurcsik, The role of the N-terminal loop in the function of the colicin E7 nuclease domain, J. Inorg. Biol. Chem. 18 (2013) 309–321.

[46] H. Schagger, Tricine–SDS–PAGE, Nat. Protoc. 1 (2006) 16–22.

[47] P. Limao-Vieira, A. Giuliani, J. Delwiche, R. Paraﬁta, R. Mota, D. Duﬂot, J.P.

Flament, E. Drage, P. Cahillane, N.J. Mason, S.V. Hoffmann, M.J. Hubin-Franskin, Acetic acid electronic state spectroscopy by high-resolution vacuum ultraviolet photo-absorption, electron impact, He(I) photoelectron spectroscopy and ab initio calculations, Chem. Phys. 324 (2006) 339–349.

[48] T. Budde, Q.A. Minta, J.A. White, A.R. Kay, Imaging free zinc in synaptic terminals in live hippocampal slices, Neuroscience 79 (1997) 347–358.

[49] T.-P. Ko, C.-C. Liao, W.-Y. Ku, K.-F. Chak, H.S. Yuan, The crystal structure of the DNase domain of colicin E7 in complex with its inhibitor Im7 protein, Structure 7 (1999) 91–102.

[50] L.M. Guzman, D. Belin, M.J. Carson, J. Beckwith, Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter, J. Bacteriol. 177 (1995) 4121–4130.

[51] K.-F. Chak, M.K. Safo, W.-Y. Ku, S.-Y. Hsieh, H.S. Yuan, The crystal structure of the immunity protein of colicin E7 suggests a possible colicin-interacting surface, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 6437–6442.

[52] C. Kállay, K.Osz, A. Dávid, Z. Valastyán, G. Malandrinos, N. Hadjiliadis, I.} Sóvágó, Zinc(II) binding ability of tri-, tetra- and penta-peptides containing two or three histidyl residues, Dalton Trans. (2007) 4040–4047.

[53] N. Sreerama, R.W. Woody, Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set, Anal. Biochem. 282 (2000) 252–260.

[54] T.R. Butt, S. Jonnalagadda, B.P. Monia, E.J. Sternberg, J.A. Marsh, J.M. Stadel, D.J.

Ecker, S.T. Crooke, Ubiquitin fusion augments the yield of cloned gene products inEscherichia coli, Proc. Natl. Acad. Sci. U.S.A. 86 (1989) 2540–2544.

[55] C. Kleanthous, D. Walker, Immunity proteins: enzyme inhibitors that avoid the active site, Trends Biochem. Sci. 26 (2001) 624–631.

[56] Sh.-y. Xu, Y.K. Gupta, Natural zinc ribbon HNH endonucleases and engineered zinc ﬁnger nicking endonuclease, Nucleic Acids Res. 41 (2013) 378–390.