Site-directed mutagenesis in a conserved motif of Epstein ... - CiteSeerX

Journal of General Virology (2003), 84, 677–686

DOI 10.1099/vir.0.18739-0

Site-directed mutagenesis in a conserved motif of Epstein–Barr virus DNase that is homologous to the catalytic centre of type II restriction endonucleases Ming-Tsan Liu,1 Hsien-Ping Hu,2 Tsuey-Ying Hsu2 and Jen-Yang Chen1,2 Correspondence Jen-Yang Chen [email protected]

Received 30 July 2002 Accepted 25 October 2002

1

National Health Research Institutes, 3F No. 109, Section 6, Min-Chuan East Road, Taipei 114, Taiwan

2

Graduate Institute of Microbiology, College of Medicine, National Taiwan University, No. 1, Section 1, Jen-Ai Road, Taipei 100, Taiwan

Sequence alignment of human herpesvirus DNases revealed that they share several conserved regions. One of these, the conserved motif D203…E225XK227 (D…EXK) in the sequence of Epstein–Barr virus (EBV) DNase, has a striking similarity to the catalytic sites of some other nucleases, including type II restriction endonucleases, l exonuclease and MutH. The predicted secondary structures of these three residues were shown to resemble the three catalytic residues of type II restriction endonucleases. Site-directed mutagenesis was carried out to replace each of the acidic residues near the motif by residues with different properties. All substitutions of D203, E225 and K227 were shown to cause significant reductions in nuclease activity. Six other acidic residues, within the conserved regions, were also replaced by Asn or Gln. Five of these six variants retained nuclease activity and mutant D195N alone lost nuclease activity. The four charged residues, D195, D203, E225 and K227, of EBV DNase were found to be important for nuclease activity. Biochemical analysis indicated that the preference for divalent cations was altered from Mg2+ to Mn2+ for mutant E225D. The DNA-binding abilities of D203E, E225D and E225Q were shown to be similar to that of wild-type. However, K227 mutants were found to have variable DNA-binding abilities: K227G and K227N mutants retained, K227E and K227D had reduced and K227R lost DNA-binding ability. Comparison of the biochemical properties of the corresponding substitutions among EBV DNase and type II restriction enzymes indicated that the D…EXK motif is most likely the putative catalytic centre of EBV DNase.

INTRODUCTION Many herpesviruses encode an alkaline deoxyribonuclease (DNase), which was originally characterized by its high pH optimum in vitro (Morrison & Keir, 1968). The DNases of various herpesviruses have similar biochemical properties, regardless of whether the enzymes are expressed in mammalian cells, Escherichia coli or using recombinant baculovirus. Biochemical studies of these herpesvirus DNases has shown that they accept both double-stranded DNA (dsDNA) and single-stranded DNA (ssDNA) as substrate and exhibit both endonuclease and exonuclease activities, a high pH optimum, an absolute requirement for divalent cations and sensitivity to high salt concentration and polyamines (Clough, 1979, 1980; Hoffmann & Cheng, 1979; Hoffmann, 1981; Baylis et al., 1989, 1991; Chen et al., 1993; Bronstein & Weber, 1996; Sheaffer et al., 1997; Kehm et al., 1998). In the case of Epstein–Barr virus (EBV) DNase, dsDNA is digested processively but ssDNA distributively (Lin et al., 1995). To digest the substrate, the enzyme first 0001-8739 G 2003 SGM

Printed in Great Britain

converts supercoiled DNA to nicked open circular DNA, then nicks open circular DNA to linear DNA and subsequently degrades the linear DNA (Hoffmann, 1981; Stolzenberg & Ooka, 1990; Lin et al., 1995). The endonuclease activity of this enzyme seems to have a structural preference but no base sequence specificity (Lin et al., 1995; Kehm et al., 1998). Exonuclease digestion of DNA is in the 59R39 direction and the products are 59monophosphate nucleosides (Lin et al., 1995; Sheaffer et al., 1997). Despite these biochemical analyses, the role(s) of the enzyme remains unclear. In the case of EBV DNase, the enzyme was induced in parallel to EBV DNA polymerase (Cheng et al., 1980) and interacted with viral DNA replication components, DNA polymerase, EA-D and mDBP (Daibata & Sairenji, 1993; Lin et al., 1995; Zeng et al., 1997). Studies of infection with a DNase null mutant of herpes simplex virus (HSV) indicated that this enzyme was required for efficient egress of capsids from the nucleus (Shao et al., 1993) and for processing non-linear or 677

M.-T. Liu and others

branched viral DNA intermediates (Martinez et al., 1996). From these results, it seems possible that herpesvirus DNases play roles in the conformational or maturational changes required for egress of DNA-containing capsids and in resolution of recombination intermediates. Herpesvirus DNases are conserved, and alignments of amino acid sequences reveal that the DNases share several conserved regions (Martinez et al., 1996; Goldstein & Weller, 1998; Liu et al., 1998; see Fig. 1). Furthermore, human cytomegalovirus (HCMV) DNase can substitute for

HSV-1 DNase and complement the growth of an HSV-1 DNase deletion mutant (Gao et al., 1998). In the absence of structural information from crystallographic studies, little is known about the cleavage mechanism and active site(s) for catalysis by herpesvirus DNases. In previous studies, we tried to locate the regions important for DNase activity by deleting amino acid residues serially from both ends of EBV DNase; a drastic loss of DNase activity was observed in mutants with small deletions and substitutions (Lin et al., 1994; Liu et al., 1998). We extended our mutational analysis to internal regions, including four conserved regions and

Fig. 1. Multiple sequence alignment and secondary structure prediction of the conserved D…EXK region of human herpesvirus DNases. (a) Sequence alignment was performed using the DIALING 2.1 program. The numbers in the sequences indicate the position of residues relative to each herpesvirus DNase. Amino acids were coloured according to the degree of conservation and amino acid properties. The acidic amino acids (D or E) conserved in all sequences are shaded red and those partially conserved in five or more are shaded in pink. The other amino acids conserved in eight or more sequences are shaded in yellow and those in less than eight are shaded green. The mutated residues and substitution amino acids are shown on the top of sequences. The aligned human herpesvirus DNases include P3HR1 and B95-8 (derived from different strains of EBV), HSV-1, HSV-2, VZV, HCMV, HHV-6, HHV-7 and HHV-8 DNases. Accession numbers for each sequence are given in Methods. The secondary structure elements are shown below the sequences: b-strands as arrows and a-helices as cylinders. The secondary structure elements of EBV DNase were predicted using the PHDsec program. (b) The secondary structure elements of EcoRI, EcoRV, PvuII, BamHI, l exonuclease and MutH are shown according to their X-ray crystal structures. The catalytic residues are shaded. 678

Journal of General Virology 84

Putative catalytic centre of EBV DNase

two non-conserved regions of herpesvirus DNases. DNAbinding and nuclease activities were abolished in all six internal deletion mutants, except that one mutant, with a deletion of residues 138–152, retained an intermediate ability to bind DNA (Liu et al., 1998). It is difficult to propose a mechanism for DNA cleavage by EBV DNase and to assign amino acid residues for an active centre(s) in catalysis simply from analyses of deletion and substitution mutations. In this study, we have tried to define functional residues by comparing conserved regions of herpesvirus DNases with other nucleases for which crystal structures have been elucidated. We identified a conserved motif, D203… E225XK227 (D…EXK), which has been shown to be a catalytic centre in some nucleases, including type II restriction endonucleases, l exonuclease and MutH (Kovall & Matthews, 1999). Analysis of the crystallographic data revealed structural homologies in the active centres, despite the lack of sequence similarities (Pingoud & Jeltsch, 1997; Kovall & Matthews, 1999). Conserved acidic residues in the D…EXK motif have been implicated in metal-ion binding, while the lysine residue is probably involved in stabilization of the charged pentacoordinate transition state or participates directly in catalysis (Pingoud & Jeltsch, 1997). In addition, the D…EXK motif was found in many nucleases by Aravind et al. (2000), using the SEALS program package. Three conserved motifs were aligned and predicted as catalytic residues in these nucleases as well as herpesvirus DNases. In order to investigate the significance of the D…EXK motif in herpesvirus DNases, site-directed mutagenesis was used to manipulate this region in EBV DNase. The results demonstrated that D203, E225 and K227 of EBV DNase play a critical role in catalysis. Based on these data, we hypothesize that herpesvirus DNases have a similar catalytic centre to the type II restriction endonucleases.

METHODS Plasmids and site-directed mutagenesis. pDNase5, a cDNA

clone from P3HR1 cells, encodes the full-length (470 amino acids) wild-type EBV DNase subcloned into pET3a (Lin et al., 1994). The mutants in this study were derived from pDNase5 and generated by site-directed mutagenesis and subcloning strategies. Site-directed mutagenesis was performed using two-step PCR mutagenesis, as described previously (Liu et al., 1998). Each mutant was generated by two consecutive PCR reactions and the corresponding regions of pDNase5 replaced with the recombinant PCR product. For example, in the construction of D195N, the two primary PCR fragments of 516 bp and 346 bp were amplified using primer pairs IDP1L and APD195N, and SPD195N and IDP1R, respectively (Table 1). The two primary PCR fragments were mixed and annealed through the sequences of APD195N and SPD195N (Table 1). The secondary PCR reaction was subsequently carried out by mixing 10 ng of each primary PCR fragment as template DNA, along with IDP1L and IDP1R as outer primers. A recombinant PCR product of 839 bp was amplified. After treatment with StuI and ApaI, the PCR-amplified fragment was used to replace the StuI–ApaI region of pDNase5. The resulting plasmid, D195N, was verified by DNA sequencing. For protein purification of wild-type and mutant DNases, the cDNA inserts were subcloned from pET3b to pET15b by digestion with http://vir.sgmjournals.org

NdeI and HindIII, as described previously (Tsai et al., 1997). The advantage of pET15b is that it carries a stretch of six histidine residues at the N terminus of the recombinant protein. Alignment of amino acid sequences of herpesvirus DNases and prediction of secondary structure. Several conserved regions

were predicted based on the alignment of amino acid sequences from different human herpesvirus DNases. Sequence alignment was performed using the DIALING 2.1 program (Morgenstern, 1999) and modified manually. P3HR1 and B95-8 DNases were derived from different EBV strains (Chen et al., 1990). Amino acid sequences of human herpesvirus 7 (HHV-7) and HHV-8 DNases were derived from their corresponding nucleotide sequences using the TRANSLATE program. Amino acid sequences of other herpesvirus DNases were obtained directly from the protein bank. Accession numbers in GenBank are: B95-8, P03217; HSV-1, P04294; HSV-2, P06489; varicella-zoster virus (VZV), P09253; HCMV, P16789; HHV-6, L14772; HHV-7, U43400 (nt 99013–100455); HHV-8, U75698 (nt 57273–58733). The secondary structure elements of EBV DNase were predicted using the PHDsec program (http://cubic.bioc.columbia.edu) and the elements of other nucleases were according to their X-ray crystal structures, including EcoRI, EcoRV, PvuII, BamHI, l exonuclease and E. coli MutH (Kim et al., 1990; Winkler et al., 1993; Cheng et al., 1994; Newman et al., 1994; Kovall & Matthews, 1997; Ban & Yang, 1998). Expression of the DNase protein in E. coli and assay for nuclease activity. Expression of pDNase5 and its mutated deriva-

tives was carried out as described previously (Lin et al., 1994). Briefly, a 2 ml culture of E. coli BL21(DE3) pLysS (Novagen) harbouring a particular plasmid was grown to exponential phase (OD600=0?4–0?6) and induced for 2 h with 1 mM IPTG. Cells were harvested and lysed with 250 ml 50 mM Tris/HCl, pH 8?0, 50 mM NaCl, 1 mM DTT and 10 % glycerol. After sonication, 5 ml of prepared lysate was assayed at various dilutions for nuclease activity (Chen et al., 1982; Lin et al., 1994). The reaction mixture contained 50 mM Tris/HCl, pH 8?0, 4 mM b-mercaptoethanol, 4 mM MgCl2 and 3 mg E. coli [14C]DNA (26104 c.p.m.). The reaction mixture was incubated at 37 ˚C for 1 h and then stopped by the addition of 25 ml sheared calf thymus DNA (2 mg ml21) followed by 25 ml 50 % trichloroacetic acid. DNase activity was defined as the amount of enzyme that converted [14C]DNA to acid-soluble material. The relative nuclease activity was normalized with the DNase protein concentration, as described previously (Liu et al., 1998). The nuclease activity of each clone minus that of cells harbouring pET3a gave the net nuclease activity. The specific nuclease activity was calculated by dividing the net activity of each clone by the intensity of each expressed protein in enhanced chemiluminescence (ECL) Western blot analysis (Tsai et al., 1997), which was quantified by scanning with a densitometer. The relative nuclease activity is the specific nuclease activity of each clone divided by that of pDNase5. Expression of recombinant EBV DNase and purification of His-tagged DNase. Purification of wild-type and mutant proteins

was carried out as follows. E. coli BL21(DE3) pLysS harbouring pET-DNase15b or its mutated derivatives was grown to exponential phase (OD600=0?4–0?6) and induced for 2 h with 1 mM IPTG. Cells were harvested by centrifugation and then pellets were resuspended in binding buffer (5 mM imidazole, 500 mM NaCl, 20 mM Tris/HCl, pH 8?0, 0?1 % NP-40 and 10 % glycerol). The lysate was sonicated and clarified by centrifugation at 39 000 g for 20 min. The supernatant was used as starting material for DNase purification. The starting material prepared above was applied to a His-Bind column (Novagen). The column was washed sequentially with 10 bed vols of TNGI-5 (20 mM Tris/HCl, pH 8?0, 500 mM NaCl, 10 % glycerol, 5 mM imidazole), 5 bed vols of TGI-30 (20 mM Tris/ HCl, pH 8?0, 10 % glycerol, 30 mM imidazole) and 5 bed vols of TGI-60 (20 mM Tris/HCl, pH 8?0, 10 % glycerol, 60 mM imidazole). 679


Table 1. Oligonucleotides primers used for PCR mutagenesis

Primer IDP1L IDP1R SPD195N APD195N SPE209Q APE209Q SPD203 APD203 SPD203N APD203N SPD213N APD213N SPD219N APD219N SPE225 APE225 SPE225Q APE225Q SPK227 APK227 SPK227E APK227E SPE238Q APE238Q SPD240N APD240N

Sequences (59R39)*

Location on BGLF5

Sense or antisense

Uses/for the construction of

AGACTGAGGCCTTTGTCC ATGACCGGGGCCCAGTTTG TAGTCCCACGaatGGCATTTTTG CAAAAATGCCattCGTGGGACTA CGTCAATGTGcagTCACAGGGAG CTCCCTGTGActgCACATTGACG GGTGTCTCTGvasCTTTGCGTCA TGACGCAAAGstbCAGAGACACC GGTGTCTCTGaatCTTTGCGTCA TGACGCAAAGattCAGAGACACC GTCACAGGGAaacTTTATACTGT ACAGTATAAAgttTCCCTGTGAC ACTGTTCACCaacCGGAGCTGCA TGCAGCTCCGgttGGTGAACAGT CTGCATTTATvawATTAAGTGCC GGCACTTAATwtbATAAATGCAG CTGCATTTATcagATTAAGTGCC GGCACTTAATctgATAAATGCAG TTATGAGATTrrmTGCCGCTTCA TGAAGCGGCAkyyAATCTCATAA TTATGAGATTgaaTGCCGCTTCA TGAAGCGGCAttcAATCTCATAA TTCCAAGTCAcagTTTGACCCCA TGGGGTCAAActgTGACTTGGAA GTCAGAGTTTaacCCCATCTACC GGTAGATGGGgttAAACTCTGAC

80–97 900–918 573–595 573–595 615–637 615–637 597–619 597–619 597–619 597–619 627–649 627–649 645–667 645–667 663–685 663–685 663–685 663–685 668–690 668–690 668–690 668–690 702–724 702–724 708–730 708–730

Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense Antisense Sense anti-sense Sense Antisense Sense Antisense Sense Antisense Sense Antisense

59 outer primer 39 outer primer D195N D195N E209Q E209Q D203K/Q/E/H D203K/Q/E/H D203N D203N D213N D213N D219N D219N E225D E225D E225Q E225Q K227D/G/N/R K227D/G/N/R K227E K227E E238Q E238Q D240N D240N

*Upper-case letters indicate the sequence in BGLF5, while lower-case letters show sequences other than BGLF5, such as mutated sites. r, A+G; k, G+T; v, C+G+A; m, A+C; y, C+T; w, A+T; b, C+G+T.

The bound proteins were then eluted with 5 bed vols of TGI-100 (20 mM Tris/HCl, pH 8?0, 10 % glycerol, 100 mM imidazole). The collected TGI-100 eluate was sequentially collected into four fractions (TGI-100-1, -2, -3, -4). The protein in each fraction was analysed by 10 % SDS-PAGE. Plasmid DNA cleavage assay. For the plasmid DNA cleavage

assay, reaction mixtures contained 125 ng of purified DNase, 0?5 mg of pEGFP-C1 plasmid DNA (Invitrogen), 50 mM Tris/HCl, pH 8?0, 4 mM b-mercaptoethanol and 4 mM MgCl2 or 4 mM MnCl2. The reactions were incubated at 37 ˚C, aliquots were taken at the indicated times and analysed by agarose gel electrophoresis. Similar assays were carried out using linear plasmid DNA, which was cut by HindIII. Expression of the DNase protein by in vitro transcription/ translation and assay of DNA-binding activity. For measuring

the DNA-binding ability of EBV DNase, [35S]methionine-labelled DNase and dsDNA cellulose chromatography were used as described previously (Liu et al., 1998). Using a TNT T7 coupled reticulocyte lysate system (Promega), DNA templates of pDNase5 and its mutants were transcribed with T7 RNA polymerase and translated in a rabbit reticulocyte lysate. Each 0?25 mg of purified plasmid DNA was used as an in vitro transcription/translation (IVT) template in a final volume of 12?5 ml. The IVT-expressed polypeptides were radiolabelled in the reaction mixture containing [35S]methionine. After 90 min of in vitro translation at 30 ˚C, the [35S]methionine-labelled 680

polypeptides were diluted 1 : 10 in buffer A (50 mM Tris/HCl, pH 8?0, 1 mM DTT, 1 mM PMSF, 10 % glycerol) containing 100 mg RNase A. An aliquot (12?5 ml) was retained as starting material and the remainder was applied to a 200 ml bed volume of dsDNA– cellulose (Sigma) in a Bio-Spin column (Bio-Rad) equilibrated in buffer A. The column was washed with 300 ml buffer A and then eluted with 400 ml step gradients of 100, 200, 300, 400 and 500 mM NaCl in buffer A. Five ml of starting material and 20 ml of eluates were mixed with SDS-PAGE sample buffer (100 mM Tris/HCl, pH 6?8, 4 % SDS, 5 % b-mercaptoethanol, 20 % glycerol, 0?05 % bromophenol blue) and analysed by 10 % SDS-PAGE and fluorography.

RESULTS Multiple sequence alignment and secondary structure prediction of the conserved D…EXK region of human herpesvirus DNases The critical amino acid residues of EBV DNase were predicted by sequence alignment with other human herpesvirus DNases. The alignment revealed that these enzymes share several conserved regions. Comparison of these conserved regions with those of other nucleases, with defined crystal structures, revealed that a conserved motif, Journal of General Virology 84


Fig. 2. Expression and relative nuclease activities of EBV DNase mutants. The bacterially expressed proteins of wild-type (wt) and mutants were detected by ECL Western blot using monoclonal antibody 311H to EBV DNase (bottom panel). The nuclease activity and protein intensity were determined as described in Methods. The nuclease activity was normalized with the DNase protein levels.

D203…E225X K227 (in the EBV DNase sequence), has a striking similarity to the catalytic sites of various nucleases, including type II restriction endonucleases, l exonuclease and MutH (Fig. 1a, b). The secondary structures of this motif were predicted and indicated that D203 was located in a loop and E225 and K227 were located in a b-sheet. The predicted secondary structures involving these three residues were shown to resemble three catalytic residues of the type II restriction endonucleases, l exonuclease and MutH (Fig. 1b). Based on these similarities, it seems possible that D203, E225 and K227 of EBV DNase are involved in catalysis. Analysis of the nuclease activity of EBV DNase mutants To investigate the possible functional roles of the three conserved residues, each was replaced by residues with different properties using site-directed mutagenesis. Residue D203 was changed to Asn, Lys, Gln, Glu or His; residue E225 was changed to either Asp or Gln and residue K227 was changed to Glu, Asp, Gly, Asn or Arg (Fig. 1a). In addition, the other acidic residues (D195, E209, D213, D219, E238 and D240) within conserved regions also were replaced by Asn or Gln (Fig. 1a). Wild-type and mutant DNases were cloned into the pET3a vector and expressed in E. coli BL21(DE3) pLysS. The recombinant EBV DNase was intact and contained no additional domains derived from the vector. The nuclease activities of wild-type, mutants and the vector control were measured. All mutants with substitutions of D203, E225 and K227 showed drastically reduced nuclease activity (Fig. 2). Five of the six other acidic residue mutants, E209Q, D213N, D219N, E238Q and D240N, retained their nuclease activities (>75 %) and only the mutant D195N lost nuclease activity (3 %) (Fig. 2). Residue http://vir.sgmjournals.org

Fig. 3. Purification of wild-type and mutant EBV DNases. Wild-type (WT), D203E, E225D, E225Q and K227E proteins were purified through a Ni-chelating column and 0?5 mg of each purified protein was analysed by 10 % SDS-PAGE and Coomassie brilliant blue staining. M indicates the protein molecular mass markers (kDa). The arrow indicates the position of the DNase protein.

D195 was not conserved in the alignment of herpesvirus DNases but a conserved Asp residue was found nearby in other herpesvirus DNases (Fig. 1a). Based on the analysis of nuclease activity, we found that four charged residues of EBV DNase, D195, D203, E225 and K227, may play important roles in catalysis. Biochemical characterization of EBV DNase mutants D203E, E225D, E225Q and K227E EBV DNase mutants were cloned into the pET15b vector and tagged with six histidines to characterize further their biochemical properties. His-tagged wild-type DNase and its mutants were expressed and purified by chromatography on a His-Bind column. The purified enzymes are shown in Fig. 3. The purified enzymes were incubated with supercoiled or linear DNA to determine the endonuclease and exonuclease activities, respectively. In addition, the effect of divalent cations was examined. In the case of His-tagged wild-type EBV DNase, a DNA cleavage pattern was observed with the appearance initially of nicked open circular DNA, then linear DNA and subsequently degraded DNA (Fig. 4a). DNA was degraded more rapidly in the presence of Mg2+ than Mn2+ (Fig. 4a). In the presence of Mg2+, Ca2+ was inhibitory for DNase activity (data not shown). These biochemical properties of His-tagged DNase were essentially the same as those described previously (Hoffman & Cheng, 1979; Lin et al., 1995). Therefore, the His-tagged DNase retains the same enzymatic characteristics as the intact DNase. His-tagged DNase mutants, as well as His-tagged wild-type DNase, were characterized with respect to a variety of reaction conditions. In the case of the D203E and K227E mutants, cleavage of supercoiled DNA or linear DNA could not be detected in the presence of Mg2+ or Mn2+ (Fig. 4b, e; Fig. 5a, b). For the E225Q mutant, cleavage of supercoiled DNA was not detected in the presence of 681


Fig. 4. Digestion of circular DNA by EBV DNase and mutants. Purified protein of wild-type (a), D203E (b), E225D (c), E225Q (d) and K227E (e) mutants was incubated with pEGFP-C1 DNA in the presence of MgCl2 or MnCl2. Samples taken at the times indicated were electrophoresed through 0?7 % agarose. OC and SC indicate the respective positions of the open circular and supercoiled forms of pEGFP-C1 DNA.

Mg2+ (Fig. 4d), whereas supercoiled DNA was converted to linear DNA, but not degraded further, in the presence of Mn2+ (Fig. 4d). Degradation of linear DNA by E225Q was not detected in the presence of Mg2+ or Mn2+ (Fig. 5a, b), indicating that E225Q retained slight endonuclease activity in the presence of Mn2+ but had lost exonuclease activity. In the case of E225D, both supercoiled and linear DNA were degraded in the presence of Mn2+ but not Mg2+(Fig. 4c; Fig. 5a, b), indicating an altered requirement for divalent cations from Mg2+ to Mn2+. Interestingly, the linear DNA was degraded more rapidly in the presence of Mn2+ by E225D than by wild-type DNase (Fig. 5b). Substitution of E225 by aspartic acid did not change the acidic residue, but decreased the length of the side chain, suggesting that residue E225 is involved in the structure of the metal binding site. 682

DNA-binding ability of DNase mutants To characterize the DNA-binding ability of DNase mutants, DNase mutant polypeptides were expressed and labelled with [35S]methionine using an IVT system. The IVTexpressed polypeptides were applied to a dsDNA–cellulose column and the strength of DNA binding was determined by elution with NaCl step gradients (Liu et al., 1998). As shown in Fig. 6, the IVT-expressed wild-type DNase bound efficiently to dsDNA–cellulose and the protein could be detected predominantly in the fractions eluted by 400 and 500 mM NaCl. The elution profiles of D203E, E225D and E225Q were similar to that of wild-type (Fig. 6). Although these mutants were defective in cleaving DNA, their DNAbinding abilities were not affected, suggesting that other changes might contribute to the loss of nuclease activity for Journal of General Virology 84


Fig. 5. Digestion of linear DNA by EBV DNase and mutants. Purified protein of wild-type EBV DNase and mutants were incubated with pEGFP-C1 DNA, which was linearized by HindIII in the presence of MgCl2 (a) or MnCl2 (b). Samples taken at the times indicated were electrophoresed through 0?7 % agarose.

these mutants. The IVT-expressed polypeptides from K227 mutants were shown to have different abilities to bind DNA. K227G and K227N mutants were detected predominantly at the 400 mM NaCl fraction and K227E and K227D were at the 200 and 300 mM NaCl fractions (Fig. 6). However, mutant K227R, which maintained the positive charge, was detected predominantly in the 100 mM NaCl fraction (Fig. 6), indicating that it had almost completely lost DNAbinding ability. One reason that K227R loses its nuclease activity may be through the impairment in DNA binding. Since K227G retained and K227R lost DNA-binding ability, the positive charge of K227 may not be essential for DNA binding and K227 may play a role in catalysis distinct from DNA binding.

DISCUSSION In common with other herpesviruses, EBV encodes a DNase that is expressed during the lytic cycle. It has been demonstrated that herpesvirus DNases have significant sequence homology and similar biochemical properties and there are, therefore, likely to be conserved catalytic residues. To identify them, we compared the conserved residues of herpesvirus DNases with the catalytic residues of other nucleases for which crystal structures have been determined. We found a conserved motif, D203…E225XK227, in EBV http://vir.sgmjournals.org

Fig. 6. DNA-binding ability of wild-type and mutants of EBV DNase. Wild-type (WT) and mutant DNase polypeptides, synthesized by IVT and labelled with [35S]methionine, were applied to dsDNA–cellulose columns and eluted with NaCl step gradients. Eluted fractions were analysed by 10 % SDSPAGE, followed by autoradiography. S, starting material; FL, flow-through; 100–500, 100–500 mM NaCl eluants.

DNase, which is homologous to the catalytic centres of some nucleases, including type II restriction endonucleases, l exonuclease and MutH (Kovall & Matthews, 1998, 1999). The D…EXK motif was also found in Holliday junction resolvases and related nucleases (Aravind et al., 2000). In addition to the D…EXK motif, a Gly/Leu-rich region was found near the first acidic residue of EBV DNase, similar to EcoRI, PvuII and BamHI (Fig. 1b). Furthermore, a cluster of hydrophobic residues between D203 and E225 of EBV DNase also paralleled EcoRI and BamHI (Fig. 1b). Prediction of the secondary structure of the domain including D203, E225 and K227 suggested that D203 was located in a loop and E225 and K227 in a b-sheet, similar to the location of catalytic residues of type II restriction enzymes (Fig. 1a, b). These similarities are strong enough to propose that D203, E225 and K227 of EBV DNase correspond to three functional residues of type II restriction enzymes. To test this hypothesis, each of the three conserved residues and flanking acidic residues were replaced by site-directed mutagenesis. Residue D203 was changed to Asn, Lys, Gln, Glu or His, residue E225 was changed to Asp or Gln, and 683


residue K227 was changed to Glu, Asp, Gly, Asn or Arg. All of these mutants exhibited significantly reduced nuclease activities (Fig. 2), underlining the importance of these three residues. Similar mutagenesis studies have been performed on EcoRI, EcoRV, BamHI and PvuII (Selent et al., 1992; Dorner & Schildkraut, 1994; Grabowski et al., 1995; Nastri et al., 1997). Mutation of any of the three corresponding residues of these restriction enzymes abolished nuclease activity. Mutant D430E of HSV-1 alkaline nuclease, corresponding to the D203E mutant of EBV DNase, lost exonuclease activity but retained endonuclease activity (Goldstein & Weller, 1998). According to crystallographic studies, the two acidic residues form a binding site for the catalytically essential metal ion and the Lys residue orients the water and assists in the nucleophile attack on the phosphorus atom (Pingound & Jeltsch, 1997; Kovall & Matthews, 1999). Corresponding amino acid residues D203 and E225 of EBV DNase were proposed to bind the metal ion. In a plasmid DNA cleavage assay, Mn2+ could activate the E225D mutant (Fig. 4c), indicating a change of preference for divalent cations from Mg2+ to Mn2+ and implying that the E225 residue is involved in the formation of the metal binding site. Because the radial size of Mn2+ is larger than Mg2+ and the side chain of Asp is shorter than that of Glu, the change in cation preference of the E225D mutant is most likely attributable to the short side chain of Asp forming a larger metal binding site, fitting Mn2+ better than Mg2+. Unlike E225D, D203E and K227E were not activated by Mn2+. Interestingly, a K92E mutant of EcoRV also became active in the presence of Mn2+ (Selent et al., 1992) and an E98Q mutant of MunI endonuclease switched its cofactor requirement from Mg2+ to Mn2+ (Lagunavicius & Siksnys, 1997). Unlike the K92E mutant of EcoRV, K227E of EBV DNase lost nuclease activity, even in the presence of Mn2+, which is similar to a K113E mutant of EcoRI (Grabowski et al., 1995). In addition to D203, E225 and K227, six other acidic residues near the D…EXK motif were also replaced by Asn or Gln. Mutational analysis showed that five of the six acidic residue mutants retained, and only mutant D195N lost, nuclease activity (Fig. 2). Although the D195 residue of EBV DNase is not conserved, a conserved Asp residue was aligned nearby in other herpesvirus DNases (Fig. 1a). The E238 residue of EBV DNase is partially conserved (Asp or Glu in the other herpesvirus DNases, shown in Fig. 1a), but the E238Q mutant retained nuclease activity, indicating that the carboxylate group of E238 was not required for catalysis. Many metal-dependent nucleases contain more than two catalytic acidic residues (Asp or Glu) to form metal binding sites at their active sites, such as type II restriction endonucleases (Pingound & Jeltsh, 1997), the 39R59 exonuclease activity of the Klenow fragment (Bernad et al., 1989), T4 RNase H and RAD2 (Mueser et al., 1996) and hFEN-1 (Shen et al., 1997). In the alignment of human herpesvirus DNases, wholly conserved acidic residues include E107, E166, D203 and E225 in the EBV sequence. The role of E166 in EBV DNase was predicted to involve 684

catalysis (Aravind et al., 2000). The residue E107 could not be predicted in the absence of homology. However, the presence of a D…(D\E)XK motif in the active sites of type II restriction endonucleases and other nucleases provided the information to predict the roles of D203 and E225. Although amino acid sequences of type II endonucleases and other nucleases did not exhibit any significant sequence identity, local structures, particularly in catalytic sites, showed striking resemblance (Kovall & Matthews, 1999). It was suggested that these nucleases with structural homology were diverged from a common ancestor (Kovall & Matthews, 1999). Furthermore, Glu113 of BamHI substitutes for the conserved Lys and the active site was characterized as a D…EXE motif (Newman et al., 1994). The (D\E)X(K\D\E) motif may exhibit a common function for nucleases and many nucleases contain such a motif in their active sites, such as the 39R59 exonucleases of prokaryotic and eukaryotic DNA polymerases, T4 RNase H and RAD2, bacterial RecJ exonucleases, hFEN-1 and Holliday junction resolvases and related nucleases (Bernad et al., 1989; Morrison et al., 1991; Mueser et al., 1996; Moser et al., 1997; Shen et al., 1997, 1998; Aravind & Koonin, 1998; Aravind et al., 2000). These similarities indicate that the catalytic (D\E)X(K\D\E) motif is conserved in the evolution of nucleases. Corresponding to this concept, NaeI was found to be a likely evolutionary bridge between DNA endonuclease and topoisomerase (Huai et al., 2000). Double-stranded DNA–cellulose chromatography was used to determine the DNA-binding characteristics of the mutants. Wild-type EBV DNase, expressed by IVT, bound efficiently to dsDNA and was eluted with 500 mM NaCl (Fig. 6). The DNA-binding characteristics of the D203E, E225D and E225Q mutants were similar to that of wild-type (Fig. 6). The D203 and E225 mutants of EBV DNase bound but failed to cleave DNA, similar to D91A and E111A of EcoRI, D74A and D90A of EcoRV, and D94A and E111A of BamHI (Selent et al., 1992; Dorner & Schildkraut, 1994; Grabowski et al., 1995). K227 mutants of EBV DNase showed variable ability to bind DNA: K227G and K227N mutants retained, K227E and K227D had reduced and K227R lost DNA-binding ability, indicating that the K227 residue may be involved in catalysis and that a positive charge for K227 is not required for DNA binding. Of the corresponding substitutions in other nucleases, K113A of EcoRI, K92A of EcoRV and K70A of PvuII lost nuclease activity but bound to DNA (Selent et al., 1992; Grabowski et al., 1995; Nastri et al., 1997), similar to the K227G mutant of EBV DNase. However, the K227R mutant of EBV DNase lost the ability to bind DNA, similar to K70R of PvuII (Nastri et al., 1997). Mutational analysis of K227 mutants of EBV DNase demonstrated that biochemical alteration of K227 mutants was similar to their counterparts in EcoRI and PvuII, underlining the similarity of these residues. Based on the studies of site-directed mutagenesis and biochemical properties of corresponding substitutions, it is suggested that the conserved motif D…EXK of herpesvirus DNases is Journal of General Virology 84


most likely the putative catalytic centre, and is homologous to those of type II restriction endonucleases.

ACKNOWLEDGEMENTS We thank Dr C.-H. Tsai and Dr M.-R. Chen for helpful discussions. We are indebted to Dr Tim J. Harrison at the Royal Free and University College Medical School, University College London, for critical reading of the manuscript. This work was supported by grants from the National Science Council (Grants NSC89-2320-B002-198 and NSC902320-B-002-135) and National Health Research Institutes, Taiwan.

REFERENCES Aravind, L. & Koonin, E. V. (1998). A novel family of predicted

phosphoesterases includes Drosophila prune protein and bacterial RecJ exonuclease. Trends Biochem Sci 23, 17–19. Aravind, L., Makarova, K. S. & Koonin, E. V. (2000). SURVEY AND

SUMMARY: Holliday junction resolvases and related nucleases: identification of new families, phyletic distribution and evolutionary trajectories. Nucleic Acids Res 28, 3417–3432. Ban, C. & Yang, W. (1998). Structural basis for MutH activation in

E. coli mismatch repair and relationship of MutH to restriction endonucleases. EMBO J 17, 1526–1534.

deoxyribonuclease, and DNA polymerase in EBV-activated Akata cells. Virology 196, 900–904. Dorner, L. F. & Schildkraut, I. (1994). Direct selection of binding proficient/catalytic deficient variants of BamHI endonuclease. Nucleic Acids Res 22, 1068–1074. Gao, M., Robertson, B. J., McCann, P. J., O’Boyle, D. R., Weller, S. K., Newcomb, W. W., Brown, J. C. & Weinheimer, S. P. (1998).

Functional conservations of the alkaline nuclease of herpes simplex type 1 and human cytomegalovirus. Virology 249, 460–470. Goldstein, J. N. & Weller, S. K. (1998). The exonuclease activity of

HSV-1 UL12 is required for in vivo function. Virology 244, 442–457. Grabowski, G., Jeltsch, A., Wolfes, H., Maass, G. & Alves, J. (1995).

Site-directed mutagenesis in the catalytic center of the restriction endonuclease EcoRI. Gene 157, 113–118. Hoffmann, P. J. (1981). Mechanism of degradation of duplex DNA by

the DNase induced by herpes simplex virus. J Virol 38, 1005–1014. Hoffmann, P. J. & Cheng, Y.-C. (1979). DNase induced after

infection of KB cells by herpes simplex virus type 1 or type 2. J Virol 32, 449–457. Huai, Q., Colandene, J. D., Chen, Y., Luo, F., Zhao, Y., Topal, M. D. & Ke, H. (2000). Crystal structure of NaeI – an evolutionary bridge

between DNA endonuclease and topoisomerase. EMBO J 19, 3110–3118. Kehm, E., Go¨ksu, M.-A., Bayer, S. & Knopf, C. W. (1998). Herpes

Baylis, S. A., Purifoy, D. J. & Littler, E. (1989). The characterization of

simplex virus type 1 DNase: functional analysis of the enzyme expressed by recombinant baculovirus. Intervirology 41, 110–119.

the EBV alkaline deoxyribonuclease cloned and expressed in E. coli. Nucleic Acids Res 17, 7609–7622.

Kim, Y., Grable, J. C., Love, R., Greene, P. J. & Rosenberg, J. M. (1990). Refinement of EcoRI endonuclease crystal structure: a revised

Baylis, S. A., Purifoy, D. J. & Littler, E. (1991). High-level expression

protein chain tracing. Science 249, 1307–1309.

of the Epstein–Barr virus alkaline deoxyribonuclease using a recombinant baculovirus: application to the diagnosis of nasopharyngeal carcinoma. Virology 181, 390–394.

Kovall, R. & Matthews, B. W. (1997). Toroidal structure of lambda-

Bernad, A., Blanco, L., Lazaro, J. M., Martin, G. & Salas, M. (1989).

A conserved 39R59 exonuclease active site in prokaryotic and eukaryotic DNA polymerase. Cell 59, 219–228. Bronstein, J. C. & Weber, P. C. (1996). Purification and characteri-

zation of herpes simplex virus type 1 alkaline exonuclease expressed in Escherichia coli. J Virol 70, 2008–2013. Chen, J.-Y., Liu, M.-Y., Lynn, T.-C. & Yang, C.-S. (1982). Antibody to

Epstein–Barr virus-specific DNase in patients with nasopharyngeal carcinoma. Chin J Microbiology and Immunology 15, 255–261. Chen, M.-R., Hsu, T.-Y., Chen, J.-Y. & Yang, C.-S. (1990). Molecular

characterization of a cDNA clone encoding the Epstein–Barr virus (EBV) DNase. J Virol Methods 29, 127–142. Chen, J.-Y., Liu, M.-Y., Hsu, T.-Y., Cho, S.-M. & Yang, C.-S. (1993).

Use of bacterially-expressed antigen for detection of antibodies to the EBV-specific deoxyribonuclease in sera from patients with nasopharyngeal carcinoma. J Virol Methods 45, 49–66. Cheng, Y.-C., Chen, J.-Y., Hoffmann, P. J. & Glaser, R. (1980).

Studies on the activity of DNase associated with the replication of the Epstein–Barr virus. Virology 100, 334–338.

exonuclease. Science 277, 1824–1827. Kovall, R. A. & Matthews, B. W. (1998). Structure, functional, and

evolutionary relationship between l-exonuclease and the type II restriction endonucleases. Proc Natl Acad Sci U S A 95, 7893–7897. Kovall, R. A. & Matthews, B. W. (1999). Type II restriction endonucleases: structural, functional and evolutionary relationships. Curr Opin Chem Biol 3, 578–583. Lagunavicius, A. & Siksnys, V. (1997). Site-directed mutagenesis of

putative active site residues of MunI restriction endonuclease: replacement of catalytically essential carboxylate residues triggers DNA binding specificity. Biochemistry 36, 11086–11092. Lin, S.-F., Lin, S.-W., Hsu, T.-Y., Liu, M.-Y., Chen, J.-Y. & Yang, C.-S. (1994). Functional analysis of the amino terminus of Epstein–Barr

virus deoxyribonuclease. Virology 199, 223–227. Lin, S.-F., Hsu, T.-Y., Liu, M.-Y., Lin, L.-S., Yang, H.-L., Chen, J.-Y. & Yang, C.-S. (1995). Characterization of Epstein–Barr virus DNase

and its interaction with the major DNA binding protein. Virology 208, 712–722. Liu, M.-T., Hsu, T.-Y., Lin, S.-F., Seow, S.-V., Liu, M.-Y., Chen, J.-Y. & Yang, C.-S. (1998). Distinct regions of EBV DNase are required for

nuclease and DNA binding activities. Virology 242, 6–13.

Cheng, X., Balendiran, K., Schildkraut, I. & Anderson, J. E. (1994).

Martinez, R., Sarisky, R. T., Weber, P. C. & Weller, S. K. (1996).

Structure of PvuII endonuclease with cognate DNA. EMBO J 13, 3927–3935.

Herpes simplex virus type 1 alkaline nuclease is required for efficient processing of viral DNA replication intermediates. J Virol 70, 2075–2085.

Clough, W. (1979). Deoxyribonuclease activity found in Epstein–Barr

virus producing lymphoblastoid cells. Biochemistry 18, 4517–4521. Clough, W. (1980). An endonuclease isolated from Epstein–

Barr virus-producing human lymphoblastoid cells. Proc Natl Acad Sci U S A 77, 6194–6198. Daibata, M. & Sairenji, T. (1993). Epstein–Barr virus (EBV) replica-

tion and expression of EA-D (BMRF1 gene product), virus-specific http://vir.sgmjournals.org

Morgenstern, B. (1999). Dialign 2: improvement of the segment-tosegment approach to multiple sequence alignment. Bioinformatics 15, 211–218. Morrison, A., Bell, J. B., Kunkel, T. A. & Sugino, A. (1991).

Eukaryotic DNA polymerase amino acid required for 39R59 exonuclease activity. Proc Natl Acad Sci U S A 88, 9473–9477. 685

M.-T. Liu and others Morrison, L. M. & Keir, H. M. (1968). A new DNA exonuclease in cells

Sheaffer, A. K., Weinheimer, S. P. & Tenney, D. J. (1997). The

infected with herpes virus: partial purification and properties of the enzyme. J Gen Virol 3, 337–347.

human cytomegalovirus UL98 gene encodes the conserved herpesvirus alkaline nuclease. J Gen Virol 78, 2953–2961.

Moser, M. J., Holley, W. R., Chatterjee, A. & Mian, I. S. (1997). The

Shen, B., Nolan, J. P., Sklar, L. A. & Park, M. S. (1997). Functional

proofreading domain of Escherichia coli DNA polymerase I and other DNA and/or RNA exonuclease domains. Nucleic Acids Res 25, 5110–5118.

analysis of point mutations in human flap endonuclease-1 active site. Nucleic Acids Res 25, 3332–3338.

Mueser, T. C., Nossal, N. G. & Hyde, C. C. (1996). Structure of

bacteriophage T4 RNase H, a 59 to 39 RNA–DNA and DNA–DNA exonuclease with sequence similarity to the RAD2 family of eukaryotic proteins. Cell 85, 1101–1112. Nastri, H. G., Evans, P. D., Walker, I. H. & Riggs, P. D. (1997).

Catalytic and DNA binding properties of PvuII restriction endonuclease mutants. J Biol Chem 272, 25761–25767. Newman, M., Strzelecka, T., Dorner, L. F., Schildkraut, I. & Aggarwal, A. A. (1994). Structure of restriction endonuclease

BamHI and its relationship to EcoRI. Nature 368, 660–664. Pingoud, A. & Jeltsch, A. (1997). Recognition and cleavage of DNA

by type-II restriction endonucleases. Eur J Biochem 246, 1–22. Selent, U., Ruter, T., Kohler, E. & 7 other authors (1992). A site-

directed mutagenesis study to identify amino acid residues involved in the catalytic function of the restriction endonuclease EcoRV. Biochemistry 31, 4808–4815. Shao, L., Rapp, L. M. & Weller, S. K. (1993). Herpes simplex virus 1

alkaline nuclease is required for efficient egress of capsids from the nucleus. Virology 196, 146–162.

686

Shen, B., Qiu, J., Hosfield, D. & Tainer, J. (1998). A flap

endonuclease homologs in archaebacteria exist as independent proteins. Trends Biochem Sci 23, 171–173. Stolzenberg, M. C. & Ooka, T. (1990). Purification and properties of Epstein–Barr virus DNase expressed in Escherichia coli. J Virol 64, 96–104. Tsai, C.-H., Liu, M.-T., Chen, M.-R., Lu, J., Yang, H.-L., Chen, J.-Y. & Yang, C.-S. (1997). Characterization of monoclonal antibodies to

the Zta and DNase proteins of Epstein–Barr virus. J Biomed Sci 4, 69–77. Winkler, F. K., Banner, D. W., Oefner, C., Tsernoglou, D., Brown, R. S., Heathman, S. P., Bryan, R. K., Martin, P. D., Petratos, K. & Wilson, K. S. (1993). The crystal structure of EcoRV endonuclease

and of its complexes with cognate and non-cognate DNA. EMBO J 12, 1781–1795. Zeng, Y., Middeldorp, J., Madjar, J. J. & Ooka, T. (1997). A major

DNA binding protein encoded by BALF2 open reading frame of Epstein–Barr virus (EBV) forms a complex with other EBV DNAbinding proteins: DNAase, EA-D, and DNA polymerase. Virology 239, 285–295.

Journal of General Virology 84