Characterization of the 3' exonuclease subunit DP1 of Methanococcus ja
http://www.100md.com
《核酸研究医学期刊》
1 Biocenter Oulu and Department of Biochemistry, PO Box 3000, FIN-90014 University of Oulu, Finland, 2 Department of Chemistry and 3 Department of Biology, University of Joensuu, PO Box 111, FIN-80101 Joensuu, Finland
*To whom correspondence should be addressed. Tel: +358 13 251 3697; Fax: +358 13 251 3590; Email: juhani.syvaoja@joensuu.fi
ABSTRACT
The B-subunits associated with the replicative DNA polymerases are conserved from Archaea to humans, whereas the corresponding catalytic subunits are not related. The latter belong to the B and D DNA polymerase families in eukaryotes and archaea, respectively. Sequence analysis places the B-subunits within the calcineurin-like phosphoesterase superfamily. Since residues implicated in metal binding and catalysis are well conserved in archaeal family D DNA polymerases, it has been hypothesized that the B-subunit could be responsible for the 3'-5' proofreading exonuclease activity of these enzymes. To test this hypothesis we expressed Methanococcus jannaschii DP1 (MjaDP1), the B-subunit of DNA polymerase D, in Escherichia coli, and demonstrate that MjaDP1 functions alone as a moderately active, thermostable, Mn2+-dependent 3'-5' exonuclease. The putative polymerase subunit DP2 is not required. The nuclease activity is strongly reduced by single amino acid mutations in the phosphoesterase domain indicating the requirement of this domain for the activity. MjaDP1 acts as a unidirectional, non-processive exonuclease preferring mispaired nucleotides and single-stranded DNA, suggesting that MjaDP1 functions as the proofreading exonuclease of archaeal family D DNA polymerase.
INTRODUCTION
All living organisms are faced with the task of keeping their genome intact. Prior to every cell division, the genetic information has to be replicated faithfully in order to avoid potentially harmful mutations. DNA polymerases (pols) are the key players in this process. Their catalytic mechanism is conserved from prokaryotes to humans. Most DNA polymerases utilize kinetic and steric mechanisms to achieve accurate incorporation of correct nucleotides (1). Many of them also possess a 3'-5' proofreading exonuclease activity hydrolyzing misincorporated nucleotides in order to improve fidelity. The loss of proofreading activity leads to a strong mutator phenotype and thereby can be lethal or increase cancer susceptibility (reviewed in 2).
DNA replication has been mainly studied in Bacteria and Eukarya. During recent years, studies have been extended to the third domain of life, the Archaea. Although they are prokaryotes, their replication machinery is more reminiscent of that found in eukaryotes (3). Compared to eukaryotes, archaeal replication is simpler as smaller genomes and fewer proteins are involved, but it also has several unique features. Some of them have obviously evolved to adapt to extreme environments.
In the archaeal branch of Euryarchaeota, chromosomal DNA replication involves two types of DNA polymerases, the family B and D enzymes (4). The family B enzymes have counterparts in bacteria (e.g. Escherichia coli pol II) and eukaryotes (e.g. pols , , ), whereas the family D DNA polymerase (pol D) is restricted to euryarchaeal species (reviewed in 5).
Pol D is composed of a small DP1 and a large DP2 subunit (6). It has been suggested, that DP2 is the catalytic subunit, whereas DP1 serves as an accessory factor (6,7). However, the interaction of the two subunits has been reported to be essential for both the polymerase and 3'-5' proofreading exonuclease activities (7,8). The strong catalytic activities of pol D suggest that it also plays a role in chromosomal DNA replication. Due to the lack of an archaeal genetic model, it is not known if both polymerase types are essential for the viability, and the division of labor between the two types of enzymes has remained unclear.
The amino acid sequence of the DP2 subunit has no homology to other DNA polymerases and with the exception of putative zinc-fingers, it does not contain common motifs typical of family B DNA polymerases (6,7,9). Surprisingly, DP1 shows homology to the non-catalytic B-subunits of the eukaryotic replicative pols , and , but also to the large calcineurin-like phosphoesterase superfamily defined by five conserved motifs (10,11). This superfamily consists of hydrolases with a common active site harboring two divalent transition metal ions involved in catalysis (12,13). However, the functions of family members are very diverse, including phosphoserine/threonine phosphatases, purple acid phosphatases, cAMP phosphodiesterases, sphingomyelin phosphodiesterases, nucleotidases, UDP-sugar hydrolases, as well as nucleases such as Mre11 or related bacterial SbcD proteins. The seven residues involved in metal coordination and catalysis are conserved in archaeal DP1 of family D DNA polymerases but disrupted in their eukaryotic counterparts (10,14). Based on this observation, it has been hypothesized that archaeal DP1 could function as a 3'-5' exonuclease (14). The absence of essential residues in the eukaryotic B-subunits would reflect their loss of exonuclease activity (14). This is consistent with experimental data, where the proofreading exonuclease activity of eukaryotic pol and is provided by the catalytic A-subunit and no enzymatic activity has been detected from any B-subunit.
In order to elucidate the role of the B-subunit in DNA replication and to test the hypothesis proposing an exonuclease activity for the B-subunit (DP1) of the family D DNA polymerase, we used Methanococcus jannaschii DP1 (MjaDP1) as a model. We demonstrate that MjaDP1 alone is a 3'-5' exonuclease with a preference for mispaired nucleotides at the 3' end. Our data indicate that MjaDP1 represents the proofreading exonuclease of the archaeal family D DNA polymerase.
MATERIALS AND METHODS
Materials
Radioactive nucleotides ATP, dCTP and dTTP, as well as all chromatographic columns including HiTrapTM chelating HP, HiTrapTM Q SepharoseTM Fast Flow, HiTrapTM DEAE SepharoseTM Fast Flow, PD10 desalting, NickTM and SuperdexTM 200 10/300 GL were from Amersham Biosciences. DNA oligonucleotides were synthesized by Sigma Genosys. The clone AMJJ55 containing the open-reading frame for M.jannaschii pol D DP1 (MjaDP1) was purchased from ATCC. Complete Mini, EDTA-free protease inhibitors were purchased from Roche Diagnostics GmbH.
Plasmid construction
The open-reading frame for MjaDP1 from the clone AMJJ55 (ATCC) and additional six C-terminal histidine codons were cloned into the pET3a (Novagen) expression vector using restriction enzyme digestions and PCR amplification. The sequence of the gene was confirmed.
Mutation of MjaDP1
Mutated MjaDP1 proteins were made using the QuikChange? site-directed mutagenesis kit (Stratagene). Mutations substituted single amino acids in the calcineurin-like phosphoesterase domain.
Bacterial expression of MjaDP1
Escherichia coli BL21 StarTM (DE3)pLysS (Invitrogen) cells harboring wild-type or mutated MjaDP1 were grown in 1 l of M9ZB medium at 37°C to an optical density of 0.6 at 600 nm. Protein expression was induced by overnight treatment with 0.5 mM IPTG at 18°C. After induction, cells were harvested, washed with PBS and rapidly frozen in liquid nitrogen. Frozen bacterial pellets were stored at –70°C.
Purification of recombinant MjaDP1
Frozen bacterial pellets from 1 l cultures containing wild-type or mutated MjaDP1 were suspended in 20 ml of lysis buffer (50 mM NaPi, pH 6.5, 0.5 M NaCl, 0.1% Triton X-100, 0.02% NaN3, 2 mM benzamidine HCl, 20 μg/ml RNase A, 25 U/ml benzonase, 200 μg/ml lysozyme, 20 mM MgCl2, Complete Mini, EDTA-free protease inhibitors). The suspension was incubated for 20 min at room temperature to lyse the cells and degrade nucleic acids. The cell lysate was clarified at 22 000 g for 30 min at 4°C. The soluble protein was heated to 80°C for 15 min to denature the E.coli proteins. After heating, the extract was incubated on ice for 20 min and clarified as above. MjaDP1 was purified with ?KTApurifier HPLC equipment (Amersham Biosciences) by applying the supernatant first to a 1 ml HiTrapTM chelating HP column charged with 100 mM NiCl2 and equilibrated with buffer A (50 mM NaPi, pH 6.5, 0.5 M NaCl, 0.02% NaN3). MjaDP1 protein was eluted with a linear gradient of 0–250 mM imidazole in buffer A. The protein containing fractions were combined and the buffer was changed by diluting 1:5 with buffer B (20 mM Tris–HCl, pH 7.5, 2% glycerol, 0.02% NaN3) adjusting the NaCl concentration to 100 mM prior to application to a 1 ml Q SepharoseTM Fast Flow anion exchanger. The proteins were eluted with a linear gradient of 100–500 mM NaCl in buffer B.
The purification of the wild-type protein was further continued by applying the pooled Q Sepharose fractions to either a 1 ml DEAE SepharoseTM Fast Flow anion exchanger or to a 24 ml SuperdexTM 200 10/300 GL gel filtration column. For the anion exchanger, the buffer was changed with a PD10 desalting column to buffer C (20 mM Tris–HCl, pH 7.5, 25 mM KCl, 10% glycerol, 0.02% NaN3). The MjaDP1 protein was eluted from DEAE Sepharose with a linear gradient of 25–500 mM KCl in buffer C. For gel filtration, protein containing fractions were concentrated using solid PEG20000 and then with ultrafiltration. The concentrated sample was applied to a Superdex 200 column and the proteins were separated in buffer D (20 mM Tris–HCl, pH 7.5, 0.3 M NaCl, 5% glycerol, 0.02% NaN3). After gel filtration, MjaDP1 was concentrated again by ultrafiltration. The purification was monitored after each step by Coomassie brilliant blue (CBB) stained SDS–PAGE gels and enzyme activity assays using dTMP-labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. Protein concentrations were determined by the Bradford assay.
In-gel digestion, mass spectrometry and N-terminal sequencing
For identification, the purified MjaDP1 protein was digested with trypsin in-gel as described (15). Tryptic peptides were analyzed with a Voyager-DETM STR matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (Applied Biosystems). For N-terminal sequencing, purified protein was blotted onto an Immobilon P-membrane (Millipore) and stained with Ponceau S. The sequence from the excised protein band was obtained with a ProciseTM 492 protein sequencer (Applied Biosystems).
DNA substrates
The poly(dA)200–500–oligo(dT)12–18 substrate for the nuclease assays was prepared by annealing poly(dA)200–500 to oligo(dT)12–18 at a 10:1 molar nucleotide ratio as described (16). Five nanomoles of poly(dA)200–500–oligo(dT)12–18 (as concentration of dNMPs) was labeled with 7 U of terminal deoxynucleotidyl transferase (MBI Fermentas) and 50 μCi of dTTP for 60 min at 37°C. The reaction was stopped by incubation at 70°C for 10 min. Finally, the substrate was precipitated with ethanol and ammonium acetate, and dissolved in 100 μl of TE buffer.
Single-stranded oligonucleotides were labeled at the 5' end with T4 polynucleotide kinase (Amersham Biosciences) and ATP. Unincorporated nucleotides were removed with a NickTM column. To form DNA duplexes, 10 pmol of labeled oligonucleotide was mixed with 10–20 pmol of unlabeled complementary strand in 20 mM Tris–HCl, pH 7.5, 5 mM MgCl2, 0.1 mM DTT and 0.01 mM EDTA. The mixture was heated to 100°C for 10 min and then allowed to anneal by slowly cooling to room temperature overnight. The resulting duplex DNA was purified through a 12% non-denaturating polyacrylamide gel in 0.5x TBE. DNA was eluted from gel slices with 100 μl of TE buffer at room temperature overnight. Gel slices were washed with 20 μl of TE buffer and the eluents were combined. The radioactivities of the substrates (Table 1) were measured by liquid scintillation (Perkin Elmer).
Table 1. DNA substratesa
The 3' end-labeled 24mer S6 substrate was prepared by annealing the 23mer oligonucleotide (5'-TCCGTAGGTA CGATCGCAGTCCG-3') to the 3' end of the 60mer (Table 1) at a ratio of 1:1 as above. After annealing, the 3' terminus was labeled by filling one nucleotide gap with the Klenow fragment (MBI Fermentas) and dCTP at room temperature for 15 min. Strands were separated by heating at 100°C for 5 min followed by electrophoresis in a 12% denaturing polyacrylamide gel. The 3' end-labeled single-strand S6 substrate was eluted in TE buffer as above.
Exonuclease assays
Exonuclease activity was measured either by using 3H-labeled poly(dA)200–500–oligo(dT)12–18 or 32P-labeled substrates S1–S13 (Table 1). The release of labeled nucleotides from the former was measured with a liquid scintillation counter. The shortening of the latter was detected by denaturing PAGE. The exonuclease assay mixtures contained either 5 pmol of poly(dA)200–500–oligo(dT)12–18 or 0.05 pmol of S1–S13 substrate and the indicated amounts of MjaDP1 in 20 μl of buffer E (50 mM HEPES–KOH, pH 7.5, 1 mg/ml BSA, 10% glycerol) supplemented with 0.1 or 5 mM MnCl2. The reactions were incubated at the given temperatures for 20 min. Assays with the poly(dA)200–500–oligo(dT)12–18 substrate were performed in duplicate. Reactions were stopped by adding EDTA to the final concentration of 12 mM followed by transfer onto Whatman DE-81 ion-exchange paper. The papers were washed three times for 5 min by gentle agitation in 0.3 M ammonium formate, pH 7.8, and once in 95% ethanol. After drying, the radioactivity was measured with a liquid scintillation counter (Perkin Elmer). The reactions containing 32P-labeled S1–S13 substrates were stopped by addition of an equal amount of stop-buffer (95% formamide, 15 mM EDTA, 0.1% BPB, 0.1% xylene cyanol) followed by denaturation at 100°C for 3 min. One out of 10 of the reaction mixtures were separated on a 14 x 16 cm 20% denaturating polyacrylamide gel in 1x TBE either at 50 V/cm for 2.5 h using a cooling unit, or at 8 V/cm for 17 h without cooling. The gels were fixed with 10% acetic acid, water and 30% methanol/3% glycerol for 15 min each. Finally, the gels were dried between gel drying films (Promega) under vacuum and exposed to Kodak BioMax MS-2 films. The bands were quantified using the ImageQuant 5.2-program (Molecular Dynamics).
RESULTS
MjaDP1 co-purifies with a nuclease activity
The presence of the calcineurin-like phosphoesterase domain in the replicative DNA polymerase associated B-subunits has lead to the hypothesis that the B-subunit is responsible for the 3'-5' exonuclease activity of the archaeal family D DNA polymerase (14). To test this hypothesis, we studied M.jannaschii DP1 (MjaDP1).
The open-reading frame of MjaDP1 was cloned into pET3a vector and the protein expressed in E.coli BL21 StarTM (DE3)pLysS. The C-terminal His-tag significantly facilitated removal of truncated forms during purification of the recombinant protein (data not shown). MjaDP1 was produced mostly as a soluble thermostable protein and appeared as the major band on SDS–PAGE after the heating step.
MjaDP1 was purified as described in Materials and Methods, and shown schematically in Figure 1A. The heated protein extract was applied to a HiTrapTM chelating HP column charged with Ni2+ ions, and MjaDP1 was eluted with a linear imidazole gradient. Fractions containing the MjaDP1 were combined and the protein was further purified with a Q SepharoseTM Fast Flow anion exchanger. As a last purification step either a DEAE SepharoseTM Fast Flow anion exchanger or a SuperdexTM 200 10/300 GL gel filtration column was used.
Figure 1. MjaDP1 co-purifies with a nuclease activity. MjaDP1 was produced as a His-tagged recombinant protein in E.coli and examined for nuclease activity. (A) The purification scheme. (B) Co-purification of a nuclease activity with MjaDP1. The purification of the MjaDP1 protein was monitored by CBB staining of SDS-PAGE gels. Nuclease activity was assayed at 65°C on dTMP-labeled poly(dA)200–500–oligo(dT)12–18, as described in Materials and Methods. Reactions included MnCl2 at a final concentration of 5 mM for the chelating Ni2+ column and 0.1 mM for the later purification steps. The combined fractions containing both the nuclease activity and MjaDP1 protein are highlighted in gray.
SDS–PAGE analysis revealed a highly purified MjaDP1 protein with a major 70 kDa band and a second faint 65 kDa band (Fig. 1B). The 70 kDa band corresponds to the calculated molecular mass of MjaDP1 and was found to be the full-length MjaDP1 when analyzed by N-terminal sequencing and mass spectrometry. The 65 kDa band was identified as a C-terminal truncated form of MjaDP1 probably due to premature translation termination or proteolysis. The gel filtration results suggest that MjaDP1 forms homodimers and provide a possible explanation for the co-purification of the truncated form with the full-length protein (data not shown). So far, we have not been able to separate the shorter form from the full-length protein.
The presence of a nuclease activity was tested after each purification step using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. As seen in Figure 1B, a manganese ion dependent nuclease activity co-eluted with MjaDP1 through all purification steps when tested at 65°C, but no activity was observed at 37°C (data not shown). The result obtained suggests that MjaDP1 is a thermostable nuclease. The exonuclease activity of MjaDP1 remained stable for several months at 4°C.
MjaDP1 is a thermostable exonuclease
The optimum temperature of the nuclease activity co- purifying with MjaDP1 was measured by using the 5' end-labeled single-stranded substrate (S5) (Fig. 2A and B). A single-stranded substrate was chosen since double-stranded DNA could partly denature at high temperatures and so influence the assay. As shown in Figure 2A, the substrate was degraded from the 3' end one nucleotide at a time indicating that MjaDP1 functioned as a 3' exonuclease, the optimum temperature being 60°C (Fig. 2B). However, the enzyme acted almost as efficiently at a temperature range from 55 to 75°C. Only minor activity was observed at temperatures lower than 45°C or higher than 85°C (data not shown). Similar results have been obtained for partially purified polymerase and exonuclease activities of M.jannaschii pol D holoenzyme (8,17).
Figure 2. MjaDP1 is a thermostable exonuclease. The temperature dependence of the MjaDP1 exonuclease was determined by incubating 3.9 pmol of purified MjaDP1 for 20 min with 5' end-labeled single-stranded S5 in the presence of 0.1 mM MnCl2. The products were separated on a 20% denaturing polyacrylamide gel and visualized by autoradiography. (A) A gel profile of the products. Oligonucleotide size markers are shown on the left. (B) A graphical representation of the temperature dependence. The amount of the products formed by MjaDP1 relative to the control reaction at identical temperature excluding MjaDP1 is presented. Control reactions were included to take into account possible spontaneous hydrolysis at elevated temperatures. The results are the means of three independent experiments. The highest relative activity is defined as 100%.
The key residues are identical in MjaDP1 and Pyrococcus furiosus Mre11
Since MjaDP1 and Mre11 belong to the same calcineurin-like phosphoesterase superfamily, the sequences of the domain were compared in order to determine if the catalytic mechanism is similar in these enzymes. Mre11 from P.furiosus was chosen for the alignment, because it is the only Mre11 protein whose structure has been determined (18). The amino acid sequence comparison revealed 62 identical residues in the phosphoesterase domain (22%) (Fig. 3A). The similarity between the two proteins is higher in the N-terminal part of the domain but significantly weaker in the C-terminal part. When looking at the positions of these residues in the crystal structure of Mre11 (PDB code 1II7 , Fig. 3B) we observed two major clusters of conserved residues (Fig. 3C). Cluster a is in the active site containing residues participating in catalysis, manganese and ligand binding. A smaller cluster b is located on the surface of the two first N-terminal helices. The residues, especially in the second helix, are relatively well conserved in all B-subunits (11). The reason for such high similarity in this area is difficult to explain. The most probable reason for this conservation is that the area interacts with some other protein, whose identity remains to be revealed.
Figure 3. The key amino acids of the calcineurin-like phosphoesterase domain are conserved in MjaDP1 and P.furiosus Mre11. (A) The amino acid sequence alignment. The identical residues between the two proteins are shown in red. The secondary structure elements of Mre11 (1II7 ) are marked with arrows (?-strands) and with dashed lines (-helices), whereas five motifs (I–V) typical for the members of the calcineurin-like phosphoesterase superfamily are in light blue under the sequences. The sequence alignment is based on the Pfam (42) alignment PF00149 of calcineurin-like phosphoesterases with manual inspection. (B) The protein backbone of Mre11 (PDB code 1II7 ) together with the side chains of identical residues. (C) The alignment of MjaDP1 and Mre11 (1II7 ) reveals two clusters (a and b) of conserved residues. Cluster a represents the catalytic site, in which two manganese ions (in green) and the ligand, deoxyadenosine monophosphate (dAMP) (in yellow) are shown. Cluster b is on the surface of the two first helices. The positions of mutated residues D374 and H421 of MjaDP1 are also indicated. The figures have been prepared using the program Setor (43).
Mutations in the calcineurin-like phosphoesterase domain abolish the exonuclease activity
To find out whether the observed exonuclease activity was due to MjaDP1, and not a contaminating protein, point mutations were introduced to the conserved calcineurin-like phosphoesterase domain of MjaDP1. Mutation sites were selected according to their ability to inactivate Mre11 and phosphatase (19,20), the proteins belonging to the same protein superfamily as MjaDP1. Residue D374 in phosphoesterase motif II and residue H421 in motif III were substituted with asparagines to produce two single amino acid mutants, respectively (Fig. 4A).
Figure 4. Mutations in the calcineurin-like phosphoesterase domain of MjaDP1 strongly reduce the exonuclease activity. (A) A schematic presentation of the MjaDP1 structure and the positions of mutations introduced. The sequences of the motifs II and III are from the DNA polymerase D subunit DP1 of M.jannaschii (M.ja.DP1), Methanobacterium thermoautotrophicum (M.th.DP1), P.furiosus (P.fu.DP1), A.fulgidus (A.fu.DP1), Mre11 of P.furiosus (P.fu.MRE11), A.fulgidus (A.fu.MRE11), M.jannaschii (M.ja.MRE11), M.thermoautotrophicum (M.th.MRE11), Homo sapiens (H.sa.MRE11), Arabidopsis thaliana (A.th.MRE11), Saccharomyces cerevisiae (S.ce.MRE11) and Rad32 of Schizosaccharomyces pombe (S.po.RAD32). The position of amino acid substitution is indicated by an arrow and boxed in the MjaDP1 sequence. The number of amino acid residues between the motifs II and III is shown in parentheses. (B) Mutant and wild-type (WT) MjaDP1 were expressed in E.coli and purified through Q Sepharose as described in Materials and Methods. The purification was monitored by CBB staining of respective Q Sepharose peak fractions in an SDS–polyacrylamide gel. (C) The nuclease activity of equal amounts of the mutant and wild-type MjaDP1 was assayed at 65°C in the presence of 0.1 mM MnCl2 using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. The activity of wild-type MjaDP1 was defined as 100%. The results are the mean values of two MjaDP1 preparations.
Mutant proteins were expressed and purified as described for the wild-type MjaDP1 (Fig. 4B). The very low level of the activity in preparations containing mutated proteins confirmed that MjaDP1 is indeed responsible for the exonuclease activity observed with the wild-type protein (Fig. 4C). Mutant D374N retained only 1.4% and mutant H421N 12.4% of wild-type activity. These results also highlight the importance of motifs II and III for the activity.
Nuclease activity is dependent on divalent manganese ions
The dependence of MjaDP1 exonuclease activity on Mn2+ or Mg2+ was measured at 65°C by using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate (Fig. 5). The exonuclease activity of MjaDP1 was dependent on 0.1–0.5 mM Mn2+, which could not be replaced by Mg2+. Other divalent cations, Ni2+ and Co2+, could also activate the enzyme to varying degrees (data not shown).
Figure 5. The nuclease activity of MjaDP1 is dependent on Mn2+. The dependence of the nuclease activity on Mn2+ and Mg2+ concentrations was measured at 65°C using 3.9 pmol of MjaDP1 and terminal transferase labeled poly(dA)200–500–oligo(dT)12–18.
MjaDP1 is a 3'-5' exonuclease that prefers 5' overhangs
The results in Figure 2A already suggested a 3' directed activity for MjaDP1. The directionality of MjaDP1 exonuclease was further studied by using single-stranded substrates labeled either at the 5' or 3' end. The 5' end-labeled substrate was shortened nucleotide by nucleotide from the 3' end producing a product ladder (Fig. 6A, left), while the radioactivity of the 3' end-labeled DNA was completely removed by MjaDP1 (Fig. 6A, right) indicating that MjaDP1 is a 3'-5' exonuclease.
Figure 6. MjaDP1 is a 3'–5' exonuclease preferring single-stranded DNA and double-stranded DNA with a 5' overhang. The products were separated on a 20% denaturing polyacrylamide gel followed by autoradiography. Oligonucleotide size markers are shown on the left and the structure of the substrate on the top of the corresponding autoradiograph. The asterisk indicates the position of the label. (A) The directionality of the MjaDP1 exonuclease was assayed on single-stranded 5' end-labeled S5 and 3' end-labeled S6. The amount of MjaDP1 from the left was 0, 0.5, 1, 2, 4 and 8 pmol. Reaction mixtures containing 0.1 mM MnCl2 were incubated at 65°C for 20 min. (B) The ability of MjaDP1 exonuclease to hydrolyze blunt-ended (S4) and double-stranded DNA with a 5' overhang (S1) was compared. The reactions were as in (A) with the exception that the incubation temperature was 50°C. (C) Quantification of the autoradiographs shown in (B) with ImageQuant 5.2 (Molecular Dynamics).
We next examined whether the MjaDP1 exonuclease could also hydrolyze double-stranded DNA. The experiment was carried out at 50°C in order to maintain complete complementarity of the double-stranded DNA. The double-stranded DNA used was found to start melting at temperatures higher than 50°C (data not shown). The substrate with a 5' protruding template strand (S1) was efficiently processed (Fig. 6B, right), whereas blunt-end (S4) DNA was a poor substrate (left). MjaDP1 was approximately 15 times more active on the substrate having a 5' overhang when compared to the blunt-end substrate (Fig. 6C) suggesting that single-stranded DNA was needed for the binding of the protein.
The gradual decrease in the quantity of products with decreasing size is indicative for a distributive action with only one nucleotide being removed after productive binding of the substrate. We noticed a relative increase of smaller products in the double-stranded substrate with a 5' overhang compared to the single-stranded substrate (Fig. 6B, right versus Fig. 6A, left). This difference points to a partially processive mechanism on the substrate with a 5' overhang that could maintain binding of the enzyme to the same substrate between successive nucleolytic cycles.
MjaDP1 does not hydrolyze substrates shorter than 16 nt
The minimum length of substrate that can be digested by MjaDP1 was determined by incubating the enzyme with the 5' end-labeled single-stranded substrates. As seen in Figure 7, MjaDP1 was able to degrade only substrates that were 16 nt or longer. From the 16mer only one nucleotide was hydrolyzed, indicating that MjaDP1 degrades substrates until they are 15 nt long. MjaDP1 processed longer substrates more efficiently than shorter ones (Fig. 7, S5 versus S10).
Figure 7. MjaDP1 exonuclease does not hydrolyze oligonucleotides shorter than 15 nt. 5' End-labeled single-stranded oligonucleotides of decreasing length were incubated with the indicated amounts of MjaDP1 in the presence of 0.1 mM MnCl2 for 20 min at 65°C. The products were separated on a 20% denaturating polyacrylamide gel and visualized by autoradiography.
MjaDP1 exonuclease prefers mispaired nucleotides at the 3' end
Earlier observations indicated that MjaDP1 prefers 5' overhangs. Next, we compared the ability of MjaDP1 to hydrolyze paired and mispaired nucleotides from the 3' end of DNA. MjaDP1 was incubated with such double-stranded DNA structures at 50°C. As presented in Figure 8A, the exonucleolytic attack on a mispaired nucleotide was more efficient than on a paired nucleotide (compare S2 and S3 to S1). Rapid removal of mispaired nucleotides took place and intermediates, with the mispaired nucleotides removed, accumulated (Fig. 8B, marked with arrow). Additionally, a natural stop-site for the enzyme was found to be present in the substrates employed (Fig. 8A, the first G). The preference for mispaired nucleotides was independent of the length of a mispaired stretch, as seen with the substrates containing 1, 4 (Fig. 8A) or 13 (data not shown) non-paired nucleotides. However, longer stretches of mispaired nucleotides were removed with a higher efficiency. The observed preference for mispaired nucleotides at the 3' end is characteristic for proofreading exonucleases and is consistent with the role of DP1 as a constituent of DNA polymerase holoenzyme acting in chromosomal DNA replication.
Figure 8. MjaDP1 exonuclease prefers mispaired nucleotides at the 3' end. (A) The 5' end-labeled oligonucleotides indicated were incubated with increasing concentrations of MjaDP1 (0, 0.5, 1, 2, 4 and 8 pmol, respectively) in the presence of 0.1 mM MnCl2 for 20 min at 50°C. The products were separated on a 20% denaturating polyacrylamide gel and visualized by autoradiography. Oligonucleotide size markers are shown on the left and the sequence on the right. (B) The rightmost lanes of the autoradiographs shown in (A) representing the highest MjaDP1 concentration (8 pmol) were quantified with ImageQuant 5.2. The intensity of a given band was compared to the summarized intensities of all bands present. The full-length substrate is indicated with an asterisk and the first identical, paired nucleotide in all three substrates with an arrow.
DISCUSSION
Although the B-subunits of replicative DNA polymerases are conserved from Archaea to humans, the catalytic subunits are not (10,11). The only common feature is the presence of two putative zinc ribbon motifs, which have been shown to be important for the interaction with the B-subunit of pol , and (21–25). In contrast, the deletion of the second zinc-finger region of Pyrococcus horikoshii pol D did not abolish the subunit interaction, but surprisingly the exonuclease was completely inactivated (26).
The seven residues of calcineurin-like phosphoesterase domain involved in metal coordination and catalysis are conserved in the B-subunits of archaeal family D DNA polymerases, whereas in eukaryotes, the important residues are dispersed (10,14). Based on this, Aravind and Koonin (10) proposed that archaeal B-subunits could possess phosphoesterase activity, whereas in eukaryotes the activity has probably been lost, although still sharing the same fold. The authors proposed that the archaeal B-subunits could hydrolyze pyrophosphate to facilitate the polymerization reaction. But since the archaeal pol D has very high proofreading activity without possessing the conventional proofreading exonuclease signature in the primary sequence, the B-subunits rather represent the proofreading exonuclease of this DNA polymerase family as shown here for M.jannaschii DP1 (MjaDP1).
Interestingly, contrary to other studies, MjaDP1 acts alone as a 3'-5' exonuclease of moderate activity. The DP2 subunit was not needed. The strong polymerase and exonuclease activities of the archaeal family D DNA polymerase of P.furiosus (7), P.horikoshii (9), Pyrococcus abyssi (14), M.jannaschii (8) and Archaeoglobus fulgidus (27) have been previously detected only in the associated DP1-DP2 holoenzyme. Only very recently, Shen et al. (28) demonstrated that the phosphoesterase domain of P.horikoshii DP1 is responsible for the Mn2+-dependent exonuclease activity. Unlike our results on M.jannaschii DP1, P.horikoshii DP1 was not sufficient to perform full catalysis alone. Besides DP1, efficient exonuclease activity was dependent on a peptide of at least 21 amino acids from the DP2 subunit. Why did the 3'-5' exonuclease activity of the DP1 remain undetected in previous studies? This may be explained by the divalent ion preference of the MjaDP1 exonuclease.
In previous studies, Mg2+ has been used exclusively when the activities of either the holoenzyme or the separate subunits have been studied (7–9,14,26,27). We found that the 3'-5' exonuclease activity of MjaDP1 was dependent on low concentrations of Mn2+ instead of Mg2+. A similar dependence on Mn2+ has been reported for Mre11 and its bacterial (SbcD) and archaeal (Mre11) homologs (29–32). Generally, the enzymes of the calcineurin-like phosphoesterase superfamily have been reported to be dependent on divalent metal ions such as Mn2+, Ni2+, Ca2+, Fe2+, Fe 3+ or Zn2+. Four conserved histidine nitrogen ligands in the structure of P.furiosus Mre11 are consistent with the preference of Mn2+ over Mg2+ (18), which is typically coordinated by all oxygen ligands (33). Hopfner and co-workers (18) found that in a crystal structure only one Mg2+ could bind to the active site, which was not sufficient for catalysis. Interestingly, the exonuclease activity of P.horikoshii DP1 was measured at such high concentrations of Mn2+ that completely inhibit M.jannashii DP1 (Fig. 5). Whether the differences between MjaDP1 and P.horikoshii DP1 (28) are due to the differences of the two species or simply due to the different experimental conditions, remains to be studied.
Consistent with the natural environment of M.jannaschii, MjaDP1 exonuclease has a high temperature optimum. Methanococcus jannaschii pol D holoenzyme has its temperature optimum at 65°C, whereas greater temperature stability has been observed for pyrococcal pol D, which is consistent with higher optimum growth temperatures of pyrococcal species (reviewed in 17).
The proofreading mechanism requires that an exonuclease fulfills the following criteria: it (i) acts unidirectionally at the 3' end, (ii) acts in a non-processive manner, (iii) releases dNMPs from its optimal single-stranded DNA substrate, (iv) excises preferentially a mispaired primer terminus and (v) interacts functionally with a polymerase to enhance fidelity of DNA synthesis (reviewed in 2,34). We demonstrate here that MjaDP1 meets the first four criteria indicating a potential function as a proofreading exonuclease of family D DNA polymerase. We show that (i) MjaDP1 is unidirectional at the 3' end and also (ii) largely non-processive, as seen by generation of the intermediate oligonucleotide products (Fig. 8A). MjaDP1 (iii) releases dNMPs from single-stranded substrates and (iv) it prefers mispaired nucleotides over paired ones. MjaDP1 exonuclease is able to remove not only one mispaired nucleotide, but also longer stretches of non-complementary nucleotides with high efficiency. Others have shown that (v) DP1 interacts with the DP2 subunit by co-purification (7,9,14) and co-immunoprecipitation (6). DP2 has been proposed to be the polymerase subunit of archaeal pol D holoenzyme although only very limited polymerase activity has been detected for this subunit (7). However, the mutation of two aspartates from the C-terminal region of DP2 abolished the polymerase activity of the holoenzyme (9) suggesting that DP2 is responsible for the polymerase activity. To fulfill all criteria for proofreading, the effect of DP1 on the fidelity of the polymerase remains to be studied.
MjaDP1 exonuclease is able to degrade substrates down to 15mers. A similar feature has been detected with human Ape1, a distant homolog of DP1, which leaves 14mers undigested (35). Ape1 possesses both endo- and exonuclease activities and has been proposed to be a proofreading exonuclease for inaccurate DNA polymerase ? (reviewed in 2). In contrast, the E.coli pol III exonuclease degrades oligonucleotides down to trimers (36). The difference between the required substrate lengths can be due to the different structures of domains responsible for the nuclease activity. Similar to our finding, P.horikoshii pol D holoenzyme has been reported to degrade partially double-stranded substrates to 13mers (26).
The editing exonuclease and polymerase activities of E.coli pol III holoenzyme are located on distinct subunits. This is in contrast to E.coli pol I and II, archaeal family B polymerases, bacteriophage polymerases, or eukaryotic pol , and (reviewed in 2). Our results indicate that also in euryarchaeal pol D the polymerase and exonuclease activities lie on separate subunits. It has been proposed that the modern-type system for double-stranded DNA replication likely evolved independently in the bacterial and archaeal/eukaryotic lineages (37). It is possible that E.coli pol III and archaeal pol D represent the two early evolutionary lines, where the domains have been later mixed to create new DNA polymerases. During evolution, the exonuclease activity has been acquired as part of the polymerase holoenzyme. If B-subunits of eukaryotic replicative DNA polymerases are generally devoid of an enzymatic activity, as suggested by sequence comparison (10,11,14), the functions of these subunits remain an open question.
Based on the study reported here and Shen et al. (26,28), it seems that DP1 has at least two functional domains. The calcineurin-like phosphoesterase domain is involved in hydrolysis of oligonucleotides (this study, 28), whereas the N-terminus with unknown domain structure regulates the exonuclease activity of the pol D holoenzyme. N-terminal truncation (amino acids 1–200) of P.horikoshii DP1 clearly enhanced the exonuclease activity of pol D holoenzyme, whereas it did not influence the polymerase activity (26). In contrast, the role of the third conserved domain, called the OB-fold (Fig. 4A) needs to be clarified. The OB-fold is a small structural motif named for its oligonucleotide/oligosaccharide binding activity (38). It has been implicated in binding to single-stranded DNA, e.g. in replication protein A (RPA) and breast cancer susceptibility protein 2 (BRCA2) (reviewed in 39–41).
Taken together, we demonstrate here that the B-subunit of M.jannaschii alone carries the proofreading exonuclease activity. It remains to be seen whether the DP2 subunit enhances the rather moderate exonuclease activity of MjaDP1 in pol D holoenzyme. The characteristics of MjaDP1 are summarized in Table 2. So far, pol D is the only DNA polymerase family reported to utilize an enzyme of the calcineurin-like phosphodiesterase superfamily as a proofreading subunit.
Table 2. Properties of MjaDP1
ACKNOWLEDGEMENTS
We thank Mari Raki, MSc, for the assistance with gel filtration and Dr P?ivi Piril? for technical help with determination of thermal denaturation of substrates.
REFERENCES
Kunkel,T.A. and Bebenek,K. (2000) DNA replication fidelity. Annu. Rev. Biochem., 69, 497–529.
Shevelev,I.V. and Hübscher,U. (2002) The 3'-5' exonucleases. Nature Rev. Mol. Cell Biol., 3, 364–375.
Edgell,D.R. and Doolittle,W.F. (1997) Archaea and the origin(s) of DNA replication proteins. Cell, 89, 995–998.
Cann,I.K.O., Ishino,S., Hayashi,I., Komori,K., Toh,H., Morikawa,K. and Ishino,Y. (1999) Functional interactions of a homolog of proliferating cell nuclear antigen with DNA polymerases in Archaea. J. Bacteriol., 181, 6591–6599.
Grabowski,B. and Kelman,Z. (2003) Archael DNA replication: eukaryal proteins in a bacterial context. Annu. Rev. Microbiol., 57, 487–516.
Cann,I.K.O., Komori,K., Toh,H., Kanai,S. and Ishino,Y. (1998) A heterodimeric DNA polymerase: evidence that members of Euryarchaeota possess a distinct DNA polymerase. Proc. Natl Acad. Sci. USA, 95, 14250–14255.
Uemori,T., Sato,Y., Kato,I., Doi,H. and Ishino,Y. (1997) A novel DNA polymerase in the hyperthermophilic archaeon, Pyrococcus furiosus: gene cloning, expression and characterization. Genes Cells, 2, 499–512.
Ishino,Y., Komori,K., Cann,I.K.O. and Koga,Y. (1998) A novel DNA polymerase family found in Archaea. J. Bacteriol., 180, 2232–2236.
Shen,Y., Musti,K., Hiramoto,M., Kikuchi,H., Kawarabayashi,Y. and Matsui,I. (2001) Invariant Asp-1122 and Asp-1124 are essential residues for polymerization catalysis of family D DNA polymerase from Pyrococcus horikoshii. J. Biol. Chem., 276, 27376–27383.
Aravind,L. and Koonin,E.V. (1998) Phosphoesterase domains associated with DNA polymerases of diverse origins. Nucleic Acids Res., 26, 3746–3752.
M?kiniemi,M., Pospiech,H., Kilpel?inen,S., Jokela,M., Vihinen,M. and Syv?oja,J.E. (1999) A novel family of DNA-polymerase-associated B subunits. Trends Biochem. Sci., 24, 14–16.
Goldberg,J., Huang,H.B., Kwon,Y.G., Greengard,P., Nairn,A.C. and Kuriyan,J. (1995) Three-dimensional structure of the catalytic subunit of protein serine/threonine phosphatase-1. Nature, 376, 745–753.
Lohse,D.L., Denu,J.M. and Dixon,J.E. (1995) Insights derived from the structures of the Ser/Thr phosphatases calcineurin and protein phosphatase 1. Structure, 3, 987–990.
Gueguen,Y., Rolland,J.-I., Lecompte,O., Azam,P., Le Romancer,G., Flament,D., Raffin,J.-P. and Dietrich,J. (2001) Characterization of two DNA polymerases from the hyperthermophilic euryarchaeon Pyrococcus abyssi. Eur. J. Biochem., 268, 5961–5969.
Hellman,U. (1997) Isolation of peptides for microsequencing by in-gel proteolytic digestion. In Kamp,R.M., Choli-Papadopoulou,T. and Wittmann-Liebold,B. (eds), Protein Structure Analysis. Preparation, Characterization and Microsequencing. Springer-Verlag, Heidelberg, Germany, pp. 97–104.
Syv?oja,J., Suomensaari,S., Nishida,C., Goldsmith,J.S., Chui,G.S.J., Jain,S. and Linn,S. (1990) DNA polymerases , and : three distinct enzymes from HeLa cells. Proc. Natl Acad. Sci. USA, 87, 6664–6668.
Ishino,Y. and Cann,I.K.O. (1998) The euryarchaeotes, a subdomain of Archaea, survive on a single DNA polymerase: fact or farce? Genes Genet. Syst., 73, 323–336.
Hopfner,K.-P., Karcher,A., Craig,L., Woo,T.T., Carney,J.P. and Tainer,J.A. (2001) Structural biochemistry and interaction architecture of the DNA double-strand break repair Mre11 nuclease and Rad50-ATPase. Cell, 105, 473–485.
Moreau,S., Ferguson,J.R. and Symington,L.S. (1999) The nuclease activity of Mre11 is required for meiosis but not for mating type switching, end joining, or telomere maintenance. Mol. Cell. Biol., 19, 556–566.
Zhuo,S., Clemens,J.C., Stone,R.L. and Dixon,J.E. (1994) Mutational analysis of a Ser/Thr phophatase. J. Biol. Chem., 269, 26234–26238.
Mizuno,T., Yamagishi,K., Miyazawa,H. and Hanaoka,F. (1999) Molecular architecture of the mouse DNA polymerase alpha-primase complex. Mol. Cell. Biol., 19, 7886–7896.
Smith,R.W.P. and Nasheuer,H.-P. (2002) Control of complex formation of DNA polymerase alpha-primase and cell-free DNA replication by the C-terminal amino acids of the largest subunit p180. FEBS Lett., 527, 143–146.
Giot,L., Chanet,R., Simon,M., Facca,C. and Faye,G. (1997) Involvement of the yeast DNA polymerase in DNA repair in vivo. Genetics, 146, 1239–1251.
Dua,R., Levy,D.L. and Campbell,J.L. (1998) Role of the putative zinc finger domain of Saccharomyces cerevisiae DNA polymerase in DNA replication and the S/M checkpoint pathway. J. Biol. Chem., 273, 30046–30055.
Dua,R., Edwards,S., Levy,D.L. and Campbell,J.L. (2000) Subunit interactions within the Saccharomyces cerevisiae DNA polymerase (pol ) complex. Demonstration of a dimeric pol . J. Biol. Chem., 275, 28816–28825.
Shen,Y., Tang,X.-F. and Matsui,I. (2003) Subunit interaction and regulation of activity through terminal domains of the family D DNA polymerase from Pyrococcus horikoshii. J. Biol. Chem., 278, 21247–21257.
Motz,M., Kober,I., Girardot,C., Loeser,E., Bauer,U., Albers,M., Moeckel,G., Minch,E., Voss,H., Kilger,C. and Koegl,M. (2002) Elucidation of an archaeal replication protein network to generate enhanced PCR enzymes. J. Biol. Chem., 277, 16179–16188.
Shen,Y., Tang,X.-F., Yokoyama,H., Matsui,E. and Matsui,I. (2004) A 21-amino acid peptide from the cysteine cluster II of the family D DNA polymerase from Pyrococcus horikoshii stimulates its nuclease activity which is Mre11-like and prefers manganese ion as the cofactor. Nucleic Acids Res., 32, 158–168.
Furuse,M., Nagase,Y., Tsubouchi,H., Murakami-Murofushi,K., Shibata,T. and Ohta,K. (1998) Distinct roles of two separable in vitro activities of yeast Mre11 in mitotic and meiotic recombination. EMBO J., 17, 6412–6425.
Trujillo,K.M., Yuan,S.-S.F., Lee,E.Y.-H.P. and Sung,P. (1998) Nuclease activities in a complex of human recombination and DNA repair factors Rad50, Mre11 and p95. J. Biol. Chem., 273, 21447–21450.
Connelly,J.C., de Leau,E.S., Okely,E.A. and Leach,D.R.F. (1997) Overexpression, purification and characterization of the SbcCD protein from Eschericia coli. J. Biol. Chem., 272, 19819–19826.
Hopfner,K.-P., Karcher,A., Shin,D., Fairley,C., Tainer,J.A. and Carney,J.P. (2000) Mre11 and Rad50 from Pyrococcus furiosus: cloning and biochemical characterization reveal an evolutionary conserved multiprotein machinery. J. Bacteriol., 182, 6036–6041.
Glusker,J.P. (1991) Structural aspects of metal liganding to functional groups in proteins. Adv. Protein Chem., 42, 1–76.
Kunkel,T.A. (1988) Exonucleolytic proofreading. Cell, 53, 837–840.
Wilson,D.M,3rd (2003) Properties of and substrate determinants for the exonuclease activity of human apurinic endonuclease Ape1. J. Mol. Biol., 330, 1027–1037.
Miller,H. and Perrino,F.W. (1996) Kinetic mechanism of the 3' 5' proofreading exonuclease of DNA polymerase III. Analysis by steady state and pre-steady state methods. Biochemistry, 35, 12919–12925.
Leipe,D.D., Aravind,L. and Koonin,E.V. (1999) Did DNA replication evolve twice independently? Nucleic Acids Res., 27, 3389–3401.
Murzin,A.G. (1993) OB(oligonucleotide/oligosaccharide binding)-fold: common structural and functional solution for non-homologous sequences. EMBO J., 12, 861–867.
Arcus,V. (2002) OB-fold domains: a snapshot of the evolution of sequence, structure and function. Curr. Opin. Struct. Biol., 12, 794–801.
Agrawal,V. and Kishan,K.V.R. (2003) OB-fold: growing bigger with functional consistency. Curr. Protein Pept. Sci., 4, 195–206.
Theobald,D.L., Mitton-Fry,R.M. and Wuttke,D.S. (2003) Nucleic acid recognition by OB-fold proteins. Annu. Rev. Biophys Biomol. Struct., 32, 115–133.
Bateman,A., Birney,E., Cerruti,L., Durbin,R., Etwiller,L., Eddy,S.R., Griffiths-Jones,S., Howe,K.L., Marshall,M. and Sonnhammer,E.L.L. (2002) The Pfam protein families database. Nucleic Acids Res., 30, 276–280.
Evans,S.V. (1993) SETOR: hardware-lighted three-dimensional solid model representations of macromolecules. J. Mol. Graph., 11, 134–138, 127–128.(Maarit Jokela1, Anitta Eskelinen1, Helmu)
*To whom correspondence should be addressed. Tel: +358 13 251 3697; Fax: +358 13 251 3590; Email: juhani.syvaoja@joensuu.fi
ABSTRACT
The B-subunits associated with the replicative DNA polymerases are conserved from Archaea to humans, whereas the corresponding catalytic subunits are not related. The latter belong to the B and D DNA polymerase families in eukaryotes and archaea, respectively. Sequence analysis places the B-subunits within the calcineurin-like phosphoesterase superfamily. Since residues implicated in metal binding and catalysis are well conserved in archaeal family D DNA polymerases, it has been hypothesized that the B-subunit could be responsible for the 3'-5' proofreading exonuclease activity of these enzymes. To test this hypothesis we expressed Methanococcus jannaschii DP1 (MjaDP1), the B-subunit of DNA polymerase D, in Escherichia coli, and demonstrate that MjaDP1 functions alone as a moderately active, thermostable, Mn2+-dependent 3'-5' exonuclease. The putative polymerase subunit DP2 is not required. The nuclease activity is strongly reduced by single amino acid mutations in the phosphoesterase domain indicating the requirement of this domain for the activity. MjaDP1 acts as a unidirectional, non-processive exonuclease preferring mispaired nucleotides and single-stranded DNA, suggesting that MjaDP1 functions as the proofreading exonuclease of archaeal family D DNA polymerase.
INTRODUCTION
All living organisms are faced with the task of keeping their genome intact. Prior to every cell division, the genetic information has to be replicated faithfully in order to avoid potentially harmful mutations. DNA polymerases (pols) are the key players in this process. Their catalytic mechanism is conserved from prokaryotes to humans. Most DNA polymerases utilize kinetic and steric mechanisms to achieve accurate incorporation of correct nucleotides (1). Many of them also possess a 3'-5' proofreading exonuclease activity hydrolyzing misincorporated nucleotides in order to improve fidelity. The loss of proofreading activity leads to a strong mutator phenotype and thereby can be lethal or increase cancer susceptibility (reviewed in 2).
DNA replication has been mainly studied in Bacteria and Eukarya. During recent years, studies have been extended to the third domain of life, the Archaea. Although they are prokaryotes, their replication machinery is more reminiscent of that found in eukaryotes (3). Compared to eukaryotes, archaeal replication is simpler as smaller genomes and fewer proteins are involved, but it also has several unique features. Some of them have obviously evolved to adapt to extreme environments.
In the archaeal branch of Euryarchaeota, chromosomal DNA replication involves two types of DNA polymerases, the family B and D enzymes (4). The family B enzymes have counterparts in bacteria (e.g. Escherichia coli pol II) and eukaryotes (e.g. pols , , ), whereas the family D DNA polymerase (pol D) is restricted to euryarchaeal species (reviewed in 5).
Pol D is composed of a small DP1 and a large DP2 subunit (6). It has been suggested, that DP2 is the catalytic subunit, whereas DP1 serves as an accessory factor (6,7). However, the interaction of the two subunits has been reported to be essential for both the polymerase and 3'-5' proofreading exonuclease activities (7,8). The strong catalytic activities of pol D suggest that it also plays a role in chromosomal DNA replication. Due to the lack of an archaeal genetic model, it is not known if both polymerase types are essential for the viability, and the division of labor between the two types of enzymes has remained unclear.
The amino acid sequence of the DP2 subunit has no homology to other DNA polymerases and with the exception of putative zinc-fingers, it does not contain common motifs typical of family B DNA polymerases (6,7,9). Surprisingly, DP1 shows homology to the non-catalytic B-subunits of the eukaryotic replicative pols , and , but also to the large calcineurin-like phosphoesterase superfamily defined by five conserved motifs (10,11). This superfamily consists of hydrolases with a common active site harboring two divalent transition metal ions involved in catalysis (12,13). However, the functions of family members are very diverse, including phosphoserine/threonine phosphatases, purple acid phosphatases, cAMP phosphodiesterases, sphingomyelin phosphodiesterases, nucleotidases, UDP-sugar hydrolases, as well as nucleases such as Mre11 or related bacterial SbcD proteins. The seven residues involved in metal coordination and catalysis are conserved in archaeal DP1 of family D DNA polymerases but disrupted in their eukaryotic counterparts (10,14). Based on this observation, it has been hypothesized that archaeal DP1 could function as a 3'-5' exonuclease (14). The absence of essential residues in the eukaryotic B-subunits would reflect their loss of exonuclease activity (14). This is consistent with experimental data, where the proofreading exonuclease activity of eukaryotic pol and is provided by the catalytic A-subunit and no enzymatic activity has been detected from any B-subunit.
In order to elucidate the role of the B-subunit in DNA replication and to test the hypothesis proposing an exonuclease activity for the B-subunit (DP1) of the family D DNA polymerase, we used Methanococcus jannaschii DP1 (MjaDP1) as a model. We demonstrate that MjaDP1 alone is a 3'-5' exonuclease with a preference for mispaired nucleotides at the 3' end. Our data indicate that MjaDP1 represents the proofreading exonuclease of the archaeal family D DNA polymerase.
MATERIALS AND METHODS
Materials
Radioactive nucleotides ATP, dCTP and dTTP, as well as all chromatographic columns including HiTrapTM chelating HP, HiTrapTM Q SepharoseTM Fast Flow, HiTrapTM DEAE SepharoseTM Fast Flow, PD10 desalting, NickTM and SuperdexTM 200 10/300 GL were from Amersham Biosciences. DNA oligonucleotides were synthesized by Sigma Genosys. The clone AMJJ55 containing the open-reading frame for M.jannaschii pol D DP1 (MjaDP1) was purchased from ATCC. Complete Mini, EDTA-free protease inhibitors were purchased from Roche Diagnostics GmbH.
Plasmid construction
The open-reading frame for MjaDP1 from the clone AMJJ55 (ATCC) and additional six C-terminal histidine codons were cloned into the pET3a (Novagen) expression vector using restriction enzyme digestions and PCR amplification. The sequence of the gene was confirmed.
Mutation of MjaDP1
Mutated MjaDP1 proteins were made using the QuikChange? site-directed mutagenesis kit (Stratagene). Mutations substituted single amino acids in the calcineurin-like phosphoesterase domain.
Bacterial expression of MjaDP1
Escherichia coli BL21 StarTM (DE3)pLysS (Invitrogen) cells harboring wild-type or mutated MjaDP1 were grown in 1 l of M9ZB medium at 37°C to an optical density of 0.6 at 600 nm. Protein expression was induced by overnight treatment with 0.5 mM IPTG at 18°C. After induction, cells were harvested, washed with PBS and rapidly frozen in liquid nitrogen. Frozen bacterial pellets were stored at –70°C.
Purification of recombinant MjaDP1
Frozen bacterial pellets from 1 l cultures containing wild-type or mutated MjaDP1 were suspended in 20 ml of lysis buffer (50 mM NaPi, pH 6.5, 0.5 M NaCl, 0.1% Triton X-100, 0.02% NaN3, 2 mM benzamidine HCl, 20 μg/ml RNase A, 25 U/ml benzonase, 200 μg/ml lysozyme, 20 mM MgCl2, Complete Mini, EDTA-free protease inhibitors). The suspension was incubated for 20 min at room temperature to lyse the cells and degrade nucleic acids. The cell lysate was clarified at 22 000 g for 30 min at 4°C. The soluble protein was heated to 80°C for 15 min to denature the E.coli proteins. After heating, the extract was incubated on ice for 20 min and clarified as above. MjaDP1 was purified with ?KTApurifier HPLC equipment (Amersham Biosciences) by applying the supernatant first to a 1 ml HiTrapTM chelating HP column charged with 100 mM NiCl2 and equilibrated with buffer A (50 mM NaPi, pH 6.5, 0.5 M NaCl, 0.02% NaN3). MjaDP1 protein was eluted with a linear gradient of 0–250 mM imidazole in buffer A. The protein containing fractions were combined and the buffer was changed by diluting 1:5 with buffer B (20 mM Tris–HCl, pH 7.5, 2% glycerol, 0.02% NaN3) adjusting the NaCl concentration to 100 mM prior to application to a 1 ml Q SepharoseTM Fast Flow anion exchanger. The proteins were eluted with a linear gradient of 100–500 mM NaCl in buffer B.
The purification of the wild-type protein was further continued by applying the pooled Q Sepharose fractions to either a 1 ml DEAE SepharoseTM Fast Flow anion exchanger or to a 24 ml SuperdexTM 200 10/300 GL gel filtration column. For the anion exchanger, the buffer was changed with a PD10 desalting column to buffer C (20 mM Tris–HCl, pH 7.5, 25 mM KCl, 10% glycerol, 0.02% NaN3). The MjaDP1 protein was eluted from DEAE Sepharose with a linear gradient of 25–500 mM KCl in buffer C. For gel filtration, protein containing fractions were concentrated using solid PEG20000 and then with ultrafiltration. The concentrated sample was applied to a Superdex 200 column and the proteins were separated in buffer D (20 mM Tris–HCl, pH 7.5, 0.3 M NaCl, 5% glycerol, 0.02% NaN3). After gel filtration, MjaDP1 was concentrated again by ultrafiltration. The purification was monitored after each step by Coomassie brilliant blue (CBB) stained SDS–PAGE gels and enzyme activity assays using dTMP-labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. Protein concentrations were determined by the Bradford assay.
In-gel digestion, mass spectrometry and N-terminal sequencing
For identification, the purified MjaDP1 protein was digested with trypsin in-gel as described (15). Tryptic peptides were analyzed with a Voyager-DETM STR matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (Applied Biosystems). For N-terminal sequencing, purified protein was blotted onto an Immobilon P-membrane (Millipore) and stained with Ponceau S. The sequence from the excised protein band was obtained with a ProciseTM 492 protein sequencer (Applied Biosystems).
DNA substrates
The poly(dA)200–500–oligo(dT)12–18 substrate for the nuclease assays was prepared by annealing poly(dA)200–500 to oligo(dT)12–18 at a 10:1 molar nucleotide ratio as described (16). Five nanomoles of poly(dA)200–500–oligo(dT)12–18 (as concentration of dNMPs) was labeled with 7 U of terminal deoxynucleotidyl transferase (MBI Fermentas) and 50 μCi of dTTP for 60 min at 37°C. The reaction was stopped by incubation at 70°C for 10 min. Finally, the substrate was precipitated with ethanol and ammonium acetate, and dissolved in 100 μl of TE buffer.
Single-stranded oligonucleotides were labeled at the 5' end with T4 polynucleotide kinase (Amersham Biosciences) and ATP. Unincorporated nucleotides were removed with a NickTM column. To form DNA duplexes, 10 pmol of labeled oligonucleotide was mixed with 10–20 pmol of unlabeled complementary strand in 20 mM Tris–HCl, pH 7.5, 5 mM MgCl2, 0.1 mM DTT and 0.01 mM EDTA. The mixture was heated to 100°C for 10 min and then allowed to anneal by slowly cooling to room temperature overnight. The resulting duplex DNA was purified through a 12% non-denaturating polyacrylamide gel in 0.5x TBE. DNA was eluted from gel slices with 100 μl of TE buffer at room temperature overnight. Gel slices were washed with 20 μl of TE buffer and the eluents were combined. The radioactivities of the substrates (Table 1) were measured by liquid scintillation (Perkin Elmer).
Table 1. DNA substratesa
The 3' end-labeled 24mer S6 substrate was prepared by annealing the 23mer oligonucleotide (5'-TCCGTAGGTA CGATCGCAGTCCG-3') to the 3' end of the 60mer (Table 1) at a ratio of 1:1 as above. After annealing, the 3' terminus was labeled by filling one nucleotide gap with the Klenow fragment (MBI Fermentas) and dCTP at room temperature for 15 min. Strands were separated by heating at 100°C for 5 min followed by electrophoresis in a 12% denaturing polyacrylamide gel. The 3' end-labeled single-strand S6 substrate was eluted in TE buffer as above.
Exonuclease assays
Exonuclease activity was measured either by using 3H-labeled poly(dA)200–500–oligo(dT)12–18 or 32P-labeled substrates S1–S13 (Table 1). The release of labeled nucleotides from the former was measured with a liquid scintillation counter. The shortening of the latter was detected by denaturing PAGE. The exonuclease assay mixtures contained either 5 pmol of poly(dA)200–500–oligo(dT)12–18 or 0.05 pmol of S1–S13 substrate and the indicated amounts of MjaDP1 in 20 μl of buffer E (50 mM HEPES–KOH, pH 7.5, 1 mg/ml BSA, 10% glycerol) supplemented with 0.1 or 5 mM MnCl2. The reactions were incubated at the given temperatures for 20 min. Assays with the poly(dA)200–500–oligo(dT)12–18 substrate were performed in duplicate. Reactions were stopped by adding EDTA to the final concentration of 12 mM followed by transfer onto Whatman DE-81 ion-exchange paper. The papers were washed three times for 5 min by gentle agitation in 0.3 M ammonium formate, pH 7.8, and once in 95% ethanol. After drying, the radioactivity was measured with a liquid scintillation counter (Perkin Elmer). The reactions containing 32P-labeled S1–S13 substrates were stopped by addition of an equal amount of stop-buffer (95% formamide, 15 mM EDTA, 0.1% BPB, 0.1% xylene cyanol) followed by denaturation at 100°C for 3 min. One out of 10 of the reaction mixtures were separated on a 14 x 16 cm 20% denaturating polyacrylamide gel in 1x TBE either at 50 V/cm for 2.5 h using a cooling unit, or at 8 V/cm for 17 h without cooling. The gels were fixed with 10% acetic acid, water and 30% methanol/3% glycerol for 15 min each. Finally, the gels were dried between gel drying films (Promega) under vacuum and exposed to Kodak BioMax MS-2 films. The bands were quantified using the ImageQuant 5.2-program (Molecular Dynamics).
RESULTS
MjaDP1 co-purifies with a nuclease activity
The presence of the calcineurin-like phosphoesterase domain in the replicative DNA polymerase associated B-subunits has lead to the hypothesis that the B-subunit is responsible for the 3'-5' exonuclease activity of the archaeal family D DNA polymerase (14). To test this hypothesis, we studied M.jannaschii DP1 (MjaDP1).
The open-reading frame of MjaDP1 was cloned into pET3a vector and the protein expressed in E.coli BL21 StarTM (DE3)pLysS. The C-terminal His-tag significantly facilitated removal of truncated forms during purification of the recombinant protein (data not shown). MjaDP1 was produced mostly as a soluble thermostable protein and appeared as the major band on SDS–PAGE after the heating step.
MjaDP1 was purified as described in Materials and Methods, and shown schematically in Figure 1A. The heated protein extract was applied to a HiTrapTM chelating HP column charged with Ni2+ ions, and MjaDP1 was eluted with a linear imidazole gradient. Fractions containing the MjaDP1 were combined and the protein was further purified with a Q SepharoseTM Fast Flow anion exchanger. As a last purification step either a DEAE SepharoseTM Fast Flow anion exchanger or a SuperdexTM 200 10/300 GL gel filtration column was used.
Figure 1. MjaDP1 co-purifies with a nuclease activity. MjaDP1 was produced as a His-tagged recombinant protein in E.coli and examined for nuclease activity. (A) The purification scheme. (B) Co-purification of a nuclease activity with MjaDP1. The purification of the MjaDP1 protein was monitored by CBB staining of SDS-PAGE gels. Nuclease activity was assayed at 65°C on dTMP-labeled poly(dA)200–500–oligo(dT)12–18, as described in Materials and Methods. Reactions included MnCl2 at a final concentration of 5 mM for the chelating Ni2+ column and 0.1 mM for the later purification steps. The combined fractions containing both the nuclease activity and MjaDP1 protein are highlighted in gray.
SDS–PAGE analysis revealed a highly purified MjaDP1 protein with a major 70 kDa band and a second faint 65 kDa band (Fig. 1B). The 70 kDa band corresponds to the calculated molecular mass of MjaDP1 and was found to be the full-length MjaDP1 when analyzed by N-terminal sequencing and mass spectrometry. The 65 kDa band was identified as a C-terminal truncated form of MjaDP1 probably due to premature translation termination or proteolysis. The gel filtration results suggest that MjaDP1 forms homodimers and provide a possible explanation for the co-purification of the truncated form with the full-length protein (data not shown). So far, we have not been able to separate the shorter form from the full-length protein.
The presence of a nuclease activity was tested after each purification step using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. As seen in Figure 1B, a manganese ion dependent nuclease activity co-eluted with MjaDP1 through all purification steps when tested at 65°C, but no activity was observed at 37°C (data not shown). The result obtained suggests that MjaDP1 is a thermostable nuclease. The exonuclease activity of MjaDP1 remained stable for several months at 4°C.
MjaDP1 is a thermostable exonuclease
The optimum temperature of the nuclease activity co- purifying with MjaDP1 was measured by using the 5' end-labeled single-stranded substrate (S5) (Fig. 2A and B). A single-stranded substrate was chosen since double-stranded DNA could partly denature at high temperatures and so influence the assay. As shown in Figure 2A, the substrate was degraded from the 3' end one nucleotide at a time indicating that MjaDP1 functioned as a 3' exonuclease, the optimum temperature being 60°C (Fig. 2B). However, the enzyme acted almost as efficiently at a temperature range from 55 to 75°C. Only minor activity was observed at temperatures lower than 45°C or higher than 85°C (data not shown). Similar results have been obtained for partially purified polymerase and exonuclease activities of M.jannaschii pol D holoenzyme (8,17).
Figure 2. MjaDP1 is a thermostable exonuclease. The temperature dependence of the MjaDP1 exonuclease was determined by incubating 3.9 pmol of purified MjaDP1 for 20 min with 5' end-labeled single-stranded S5 in the presence of 0.1 mM MnCl2. The products were separated on a 20% denaturing polyacrylamide gel and visualized by autoradiography. (A) A gel profile of the products. Oligonucleotide size markers are shown on the left. (B) A graphical representation of the temperature dependence. The amount of the products formed by MjaDP1 relative to the control reaction at identical temperature excluding MjaDP1 is presented. Control reactions were included to take into account possible spontaneous hydrolysis at elevated temperatures. The results are the means of three independent experiments. The highest relative activity is defined as 100%.
The key residues are identical in MjaDP1 and Pyrococcus furiosus Mre11
Since MjaDP1 and Mre11 belong to the same calcineurin-like phosphoesterase superfamily, the sequences of the domain were compared in order to determine if the catalytic mechanism is similar in these enzymes. Mre11 from P.furiosus was chosen for the alignment, because it is the only Mre11 protein whose structure has been determined (18). The amino acid sequence comparison revealed 62 identical residues in the phosphoesterase domain (22%) (Fig. 3A). The similarity between the two proteins is higher in the N-terminal part of the domain but significantly weaker in the C-terminal part. When looking at the positions of these residues in the crystal structure of Mre11 (PDB code 1II7 , Fig. 3B) we observed two major clusters of conserved residues (Fig. 3C). Cluster a is in the active site containing residues participating in catalysis, manganese and ligand binding. A smaller cluster b is located on the surface of the two first N-terminal helices. The residues, especially in the second helix, are relatively well conserved in all B-subunits (11). The reason for such high similarity in this area is difficult to explain. The most probable reason for this conservation is that the area interacts with some other protein, whose identity remains to be revealed.
Figure 3. The key amino acids of the calcineurin-like phosphoesterase domain are conserved in MjaDP1 and P.furiosus Mre11. (A) The amino acid sequence alignment. The identical residues between the two proteins are shown in red. The secondary structure elements of Mre11 (1II7 ) are marked with arrows (?-strands) and with dashed lines (-helices), whereas five motifs (I–V) typical for the members of the calcineurin-like phosphoesterase superfamily are in light blue under the sequences. The sequence alignment is based on the Pfam (42) alignment PF00149 of calcineurin-like phosphoesterases with manual inspection. (B) The protein backbone of Mre11 (PDB code 1II7 ) together with the side chains of identical residues. (C) The alignment of MjaDP1 and Mre11 (1II7 ) reveals two clusters (a and b) of conserved residues. Cluster a represents the catalytic site, in which two manganese ions (in green) and the ligand, deoxyadenosine monophosphate (dAMP) (in yellow) are shown. Cluster b is on the surface of the two first helices. The positions of mutated residues D374 and H421 of MjaDP1 are also indicated. The figures have been prepared using the program Setor (43).
Mutations in the calcineurin-like phosphoesterase domain abolish the exonuclease activity
To find out whether the observed exonuclease activity was due to MjaDP1, and not a contaminating protein, point mutations were introduced to the conserved calcineurin-like phosphoesterase domain of MjaDP1. Mutation sites were selected according to their ability to inactivate Mre11 and phosphatase (19,20), the proteins belonging to the same protein superfamily as MjaDP1. Residue D374 in phosphoesterase motif II and residue H421 in motif III were substituted with asparagines to produce two single amino acid mutants, respectively (Fig. 4A).
Figure 4. Mutations in the calcineurin-like phosphoesterase domain of MjaDP1 strongly reduce the exonuclease activity. (A) A schematic presentation of the MjaDP1 structure and the positions of mutations introduced. The sequences of the motifs II and III are from the DNA polymerase D subunit DP1 of M.jannaschii (M.ja.DP1), Methanobacterium thermoautotrophicum (M.th.DP1), P.furiosus (P.fu.DP1), A.fulgidus (A.fu.DP1), Mre11 of P.furiosus (P.fu.MRE11), A.fulgidus (A.fu.MRE11), M.jannaschii (M.ja.MRE11), M.thermoautotrophicum (M.th.MRE11), Homo sapiens (H.sa.MRE11), Arabidopsis thaliana (A.th.MRE11), Saccharomyces cerevisiae (S.ce.MRE11) and Rad32 of Schizosaccharomyces pombe (S.po.RAD32). The position of amino acid substitution is indicated by an arrow and boxed in the MjaDP1 sequence. The number of amino acid residues between the motifs II and III is shown in parentheses. (B) Mutant and wild-type (WT) MjaDP1 were expressed in E.coli and purified through Q Sepharose as described in Materials and Methods. The purification was monitored by CBB staining of respective Q Sepharose peak fractions in an SDS–polyacrylamide gel. (C) The nuclease activity of equal amounts of the mutant and wild-type MjaDP1 was assayed at 65°C in the presence of 0.1 mM MnCl2 using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate. The activity of wild-type MjaDP1 was defined as 100%. The results are the mean values of two MjaDP1 preparations.
Mutant proteins were expressed and purified as described for the wild-type MjaDP1 (Fig. 4B). The very low level of the activity in preparations containing mutated proteins confirmed that MjaDP1 is indeed responsible for the exonuclease activity observed with the wild-type protein (Fig. 4C). Mutant D374N retained only 1.4% and mutant H421N 12.4% of wild-type activity. These results also highlight the importance of motifs II and III for the activity.
Nuclease activity is dependent on divalent manganese ions
The dependence of MjaDP1 exonuclease activity on Mn2+ or Mg2+ was measured at 65°C by using terminal transferase labeled poly(dA)200–500–oligo(dT)12–18 as a substrate (Fig. 5). The exonuclease activity of MjaDP1 was dependent on 0.1–0.5 mM Mn2+, which could not be replaced by Mg2+. Other divalent cations, Ni2+ and Co2+, could also activate the enzyme to varying degrees (data not shown).
Figure 5. The nuclease activity of MjaDP1 is dependent on Mn2+. The dependence of the nuclease activity on Mn2+ and Mg2+ concentrations was measured at 65°C using 3.9 pmol of MjaDP1 and terminal transferase labeled poly(dA)200–500–oligo(dT)12–18.
MjaDP1 is a 3'-5' exonuclease that prefers 5' overhangs
The results in Figure 2A already suggested a 3' directed activity for MjaDP1. The directionality of MjaDP1 exonuclease was further studied by using single-stranded substrates labeled either at the 5' or 3' end. The 5' end-labeled substrate was shortened nucleotide by nucleotide from the 3' end producing a product ladder (Fig. 6A, left), while the radioactivity of the 3' end-labeled DNA was completely removed by MjaDP1 (Fig. 6A, right) indicating that MjaDP1 is a 3'-5' exonuclease.
Figure 6. MjaDP1 is a 3'–5' exonuclease preferring single-stranded DNA and double-stranded DNA with a 5' overhang. The products were separated on a 20% denaturing polyacrylamide gel followed by autoradiography. Oligonucleotide size markers are shown on the left and the structure of the substrate on the top of the corresponding autoradiograph. The asterisk indicates the position of the label. (A) The directionality of the MjaDP1 exonuclease was assayed on single-stranded 5' end-labeled S5 and 3' end-labeled S6. The amount of MjaDP1 from the left was 0, 0.5, 1, 2, 4 and 8 pmol. Reaction mixtures containing 0.1 mM MnCl2 were incubated at 65°C for 20 min. (B) The ability of MjaDP1 exonuclease to hydrolyze blunt-ended (S4) and double-stranded DNA with a 5' overhang (S1) was compared. The reactions were as in (A) with the exception that the incubation temperature was 50°C. (C) Quantification of the autoradiographs shown in (B) with ImageQuant 5.2 (Molecular Dynamics).
We next examined whether the MjaDP1 exonuclease could also hydrolyze double-stranded DNA. The experiment was carried out at 50°C in order to maintain complete complementarity of the double-stranded DNA. The double-stranded DNA used was found to start melting at temperatures higher than 50°C (data not shown). The substrate with a 5' protruding template strand (S1) was efficiently processed (Fig. 6B, right), whereas blunt-end (S4) DNA was a poor substrate (left). MjaDP1 was approximately 15 times more active on the substrate having a 5' overhang when compared to the blunt-end substrate (Fig. 6C) suggesting that single-stranded DNA was needed for the binding of the protein.
The gradual decrease in the quantity of products with decreasing size is indicative for a distributive action with only one nucleotide being removed after productive binding of the substrate. We noticed a relative increase of smaller products in the double-stranded substrate with a 5' overhang compared to the single-stranded substrate (Fig. 6B, right versus Fig. 6A, left). This difference points to a partially processive mechanism on the substrate with a 5' overhang that could maintain binding of the enzyme to the same substrate between successive nucleolytic cycles.
MjaDP1 does not hydrolyze substrates shorter than 16 nt
The minimum length of substrate that can be digested by MjaDP1 was determined by incubating the enzyme with the 5' end-labeled single-stranded substrates. As seen in Figure 7, MjaDP1 was able to degrade only substrates that were 16 nt or longer. From the 16mer only one nucleotide was hydrolyzed, indicating that MjaDP1 degrades substrates until they are 15 nt long. MjaDP1 processed longer substrates more efficiently than shorter ones (Fig. 7, S5 versus S10).
Figure 7. MjaDP1 exonuclease does not hydrolyze oligonucleotides shorter than 15 nt. 5' End-labeled single-stranded oligonucleotides of decreasing length were incubated with the indicated amounts of MjaDP1 in the presence of 0.1 mM MnCl2 for 20 min at 65°C. The products were separated on a 20% denaturating polyacrylamide gel and visualized by autoradiography.
MjaDP1 exonuclease prefers mispaired nucleotides at the 3' end
Earlier observations indicated that MjaDP1 prefers 5' overhangs. Next, we compared the ability of MjaDP1 to hydrolyze paired and mispaired nucleotides from the 3' end of DNA. MjaDP1 was incubated with such double-stranded DNA structures at 50°C. As presented in Figure 8A, the exonucleolytic attack on a mispaired nucleotide was more efficient than on a paired nucleotide (compare S2 and S3 to S1). Rapid removal of mispaired nucleotides took place and intermediates, with the mispaired nucleotides removed, accumulated (Fig. 8B, marked with arrow). Additionally, a natural stop-site for the enzyme was found to be present in the substrates employed (Fig. 8A, the first G). The preference for mispaired nucleotides was independent of the length of a mispaired stretch, as seen with the substrates containing 1, 4 (Fig. 8A) or 13 (data not shown) non-paired nucleotides. However, longer stretches of mispaired nucleotides were removed with a higher efficiency. The observed preference for mispaired nucleotides at the 3' end is characteristic for proofreading exonucleases and is consistent with the role of DP1 as a constituent of DNA polymerase holoenzyme acting in chromosomal DNA replication.
Figure 8. MjaDP1 exonuclease prefers mispaired nucleotides at the 3' end. (A) The 5' end-labeled oligonucleotides indicated were incubated with increasing concentrations of MjaDP1 (0, 0.5, 1, 2, 4 and 8 pmol, respectively) in the presence of 0.1 mM MnCl2 for 20 min at 50°C. The products were separated on a 20% denaturating polyacrylamide gel and visualized by autoradiography. Oligonucleotide size markers are shown on the left and the sequence on the right. (B) The rightmost lanes of the autoradiographs shown in (A) representing the highest MjaDP1 concentration (8 pmol) were quantified with ImageQuant 5.2. The intensity of a given band was compared to the summarized intensities of all bands present. The full-length substrate is indicated with an asterisk and the first identical, paired nucleotide in all three substrates with an arrow.
DISCUSSION
Although the B-subunits of replicative DNA polymerases are conserved from Archaea to humans, the catalytic subunits are not (10,11). The only common feature is the presence of two putative zinc ribbon motifs, which have been shown to be important for the interaction with the B-subunit of pol , and (21–25). In contrast, the deletion of the second zinc-finger region of Pyrococcus horikoshii pol D did not abolish the subunit interaction, but surprisingly the exonuclease was completely inactivated (26).
The seven residues of calcineurin-like phosphoesterase domain involved in metal coordination and catalysis are conserved in the B-subunits of archaeal family D DNA polymerases, whereas in eukaryotes, the important residues are dispersed (10,14). Based on this, Aravind and Koonin (10) proposed that archaeal B-subunits could possess phosphoesterase activity, whereas in eukaryotes the activity has probably been lost, although still sharing the same fold. The authors proposed that the archaeal B-subunits could hydrolyze pyrophosphate to facilitate the polymerization reaction. But since the archaeal pol D has very high proofreading activity without possessing the conventional proofreading exonuclease signature in the primary sequence, the B-subunits rather represent the proofreading exonuclease of this DNA polymerase family as shown here for M.jannaschii DP1 (MjaDP1).
Interestingly, contrary to other studies, MjaDP1 acts alone as a 3'-5' exonuclease of moderate activity. The DP2 subunit was not needed. The strong polymerase and exonuclease activities of the archaeal family D DNA polymerase of P.furiosus (7), P.horikoshii (9), Pyrococcus abyssi (14), M.jannaschii (8) and Archaeoglobus fulgidus (27) have been previously detected only in the associated DP1-DP2 holoenzyme. Only very recently, Shen et al. (28) demonstrated that the phosphoesterase domain of P.horikoshii DP1 is responsible for the Mn2+-dependent exonuclease activity. Unlike our results on M.jannaschii DP1, P.horikoshii DP1 was not sufficient to perform full catalysis alone. Besides DP1, efficient exonuclease activity was dependent on a peptide of at least 21 amino acids from the DP2 subunit. Why did the 3'-5' exonuclease activity of the DP1 remain undetected in previous studies? This may be explained by the divalent ion preference of the MjaDP1 exonuclease.
In previous studies, Mg2+ has been used exclusively when the activities of either the holoenzyme or the separate subunits have been studied (7–9,14,26,27). We found that the 3'-5' exonuclease activity of MjaDP1 was dependent on low concentrations of Mn2+ instead of Mg2+. A similar dependence on Mn2+ has been reported for Mre11 and its bacterial (SbcD) and archaeal (Mre11) homologs (29–32). Generally, the enzymes of the calcineurin-like phosphoesterase superfamily have been reported to be dependent on divalent metal ions such as Mn2+, Ni2+, Ca2+, Fe2+, Fe 3+ or Zn2+. Four conserved histidine nitrogen ligands in the structure of P.furiosus Mre11 are consistent with the preference of Mn2+ over Mg2+ (18), which is typically coordinated by all oxygen ligands (33). Hopfner and co-workers (18) found that in a crystal structure only one Mg2+ could bind to the active site, which was not sufficient for catalysis. Interestingly, the exonuclease activity of P.horikoshii DP1 was measured at such high concentrations of Mn2+ that completely inhibit M.jannashii DP1 (Fig. 5). Whether the differences between MjaDP1 and P.horikoshii DP1 (28) are due to the differences of the two species or simply due to the different experimental conditions, remains to be studied.
Consistent with the natural environment of M.jannaschii, MjaDP1 exonuclease has a high temperature optimum. Methanococcus jannaschii pol D holoenzyme has its temperature optimum at 65°C, whereas greater temperature stability has been observed for pyrococcal pol D, which is consistent with higher optimum growth temperatures of pyrococcal species (reviewed in 17).
The proofreading mechanism requires that an exonuclease fulfills the following criteria: it (i) acts unidirectionally at the 3' end, (ii) acts in a non-processive manner, (iii) releases dNMPs from its optimal single-stranded DNA substrate, (iv) excises preferentially a mispaired primer terminus and (v) interacts functionally with a polymerase to enhance fidelity of DNA synthesis (reviewed in 2,34). We demonstrate here that MjaDP1 meets the first four criteria indicating a potential function as a proofreading exonuclease of family D DNA polymerase. We show that (i) MjaDP1 is unidirectional at the 3' end and also (ii) largely non-processive, as seen by generation of the intermediate oligonucleotide products (Fig. 8A). MjaDP1 (iii) releases dNMPs from single-stranded substrates and (iv) it prefers mispaired nucleotides over paired ones. MjaDP1 exonuclease is able to remove not only one mispaired nucleotide, but also longer stretches of non-complementary nucleotides with high efficiency. Others have shown that (v) DP1 interacts with the DP2 subunit by co-purification (7,9,14) and co-immunoprecipitation (6). DP2 has been proposed to be the polymerase subunit of archaeal pol D holoenzyme although only very limited polymerase activity has been detected for this subunit (7). However, the mutation of two aspartates from the C-terminal region of DP2 abolished the polymerase activity of the holoenzyme (9) suggesting that DP2 is responsible for the polymerase activity. To fulfill all criteria for proofreading, the effect of DP1 on the fidelity of the polymerase remains to be studied.
MjaDP1 exonuclease is able to degrade substrates down to 15mers. A similar feature has been detected with human Ape1, a distant homolog of DP1, which leaves 14mers undigested (35). Ape1 possesses both endo- and exonuclease activities and has been proposed to be a proofreading exonuclease for inaccurate DNA polymerase ? (reviewed in 2). In contrast, the E.coli pol III exonuclease degrades oligonucleotides down to trimers (36). The difference between the required substrate lengths can be due to the different structures of domains responsible for the nuclease activity. Similar to our finding, P.horikoshii pol D holoenzyme has been reported to degrade partially double-stranded substrates to 13mers (26).
The editing exonuclease and polymerase activities of E.coli pol III holoenzyme are located on distinct subunits. This is in contrast to E.coli pol I and II, archaeal family B polymerases, bacteriophage polymerases, or eukaryotic pol , and (reviewed in 2). Our results indicate that also in euryarchaeal pol D the polymerase and exonuclease activities lie on separate subunits. It has been proposed that the modern-type system for double-stranded DNA replication likely evolved independently in the bacterial and archaeal/eukaryotic lineages (37). It is possible that E.coli pol III and archaeal pol D represent the two early evolutionary lines, where the domains have been later mixed to create new DNA polymerases. During evolution, the exonuclease activity has been acquired as part of the polymerase holoenzyme. If B-subunits of eukaryotic replicative DNA polymerases are generally devoid of an enzymatic activity, as suggested by sequence comparison (10,11,14), the functions of these subunits remain an open question.
Based on the study reported here and Shen et al. (26,28), it seems that DP1 has at least two functional domains. The calcineurin-like phosphoesterase domain is involved in hydrolysis of oligonucleotides (this study, 28), whereas the N-terminus with unknown domain structure regulates the exonuclease activity of the pol D holoenzyme. N-terminal truncation (amino acids 1–200) of P.horikoshii DP1 clearly enhanced the exonuclease activity of pol D holoenzyme, whereas it did not influence the polymerase activity (26). In contrast, the role of the third conserved domain, called the OB-fold (Fig. 4A) needs to be clarified. The OB-fold is a small structural motif named for its oligonucleotide/oligosaccharide binding activity (38). It has been implicated in binding to single-stranded DNA, e.g. in replication protein A (RPA) and breast cancer susceptibility protein 2 (BRCA2) (reviewed in 39–41).
Taken together, we demonstrate here that the B-subunit of M.jannaschii alone carries the proofreading exonuclease activity. It remains to be seen whether the DP2 subunit enhances the rather moderate exonuclease activity of MjaDP1 in pol D holoenzyme. The characteristics of MjaDP1 are summarized in Table 2. So far, pol D is the only DNA polymerase family reported to utilize an enzyme of the calcineurin-like phosphodiesterase superfamily as a proofreading subunit.
Table 2. Properties of MjaDP1
ACKNOWLEDGEMENTS
We thank Mari Raki, MSc, for the assistance with gel filtration and Dr P?ivi Piril? for technical help with determination of thermal denaturation of substrates.
REFERENCES
Kunkel,T.A. and Bebenek,K. (2000) DNA replication fidelity. Annu. Rev. Biochem., 69, 497–529.
Shevelev,I.V. and Hübscher,U. (2002) The 3'-5' exonucleases. Nature Rev. Mol. Cell Biol., 3, 364–375.
Edgell,D.R. and Doolittle,W.F. (1997) Archaea and the origin(s) of DNA replication proteins. Cell, 89, 995–998.
Cann,I.K.O., Ishino,S., Hayashi,I., Komori,K., Toh,H., Morikawa,K. and Ishino,Y. (1999) Functional interactions of a homolog of proliferating cell nuclear antigen with DNA polymerases in Archaea. J. Bacteriol., 181, 6591–6599.
Grabowski,B. and Kelman,Z. (2003) Archael DNA replication: eukaryal proteins in a bacterial context. Annu. Rev. Microbiol., 57, 487–516.
Cann,I.K.O., Komori,K., Toh,H., Kanai,S. and Ishino,Y. (1998) A heterodimeric DNA polymerase: evidence that members of Euryarchaeota possess a distinct DNA polymerase. Proc. Natl Acad. Sci. USA, 95, 14250–14255.
Uemori,T., Sato,Y., Kato,I., Doi,H. and Ishino,Y. (1997) A novel DNA polymerase in the hyperthermophilic archaeon, Pyrococcus furiosus: gene cloning, expression and characterization. Genes Cells, 2, 499–512.
Ishino,Y., Komori,K., Cann,I.K.O. and Koga,Y. (1998) A novel DNA polymerase family found in Archaea. J. Bacteriol., 180, 2232–2236.
Shen,Y., Musti,K., Hiramoto,M., Kikuchi,H., Kawarabayashi,Y. and Matsui,I. (2001) Invariant Asp-1122 and Asp-1124 are essential residues for polymerization catalysis of family D DNA polymerase from Pyrococcus horikoshii. J. Biol. Chem., 276, 27376–27383.
Aravind,L. and Koonin,E.V. (1998) Phosphoesterase domains associated with DNA polymerases of diverse origins. Nucleic Acids Res., 26, 3746–3752.
M?kiniemi,M., Pospiech,H., Kilpel?inen,S., Jokela,M., Vihinen,M. and Syv?oja,J.E. (1999) A novel family of DNA-polymerase-associated B subunits. Trends Biochem. Sci., 24, 14–16.
Goldberg,J., Huang,H.B., Kwon,Y.G., Greengard,P., Nairn,A.C. and Kuriyan,J. (1995) Three-dimensional structure of the catalytic subunit of protein serine/threonine phosphatase-1. Nature, 376, 745–753.
Lohse,D.L., Denu,J.M. and Dixon,J.E. (1995) Insights derived from the structures of the Ser/Thr phosphatases calcineurin and protein phosphatase 1. Structure, 3, 987–990.
Gueguen,Y., Rolland,J.-I., Lecompte,O., Azam,P., Le Romancer,G., Flament,D., Raffin,J.-P. and Dietrich,J. (2001) Characterization of two DNA polymerases from the hyperthermophilic euryarchaeon Pyrococcus abyssi. Eur. J. Biochem., 268, 5961–5969.
Hellman,U. (1997) Isolation of peptides for microsequencing by in-gel proteolytic digestion. In Kamp,R.M., Choli-Papadopoulou,T. and Wittmann-Liebold,B. (eds), Protein Structure Analysis. Preparation, Characterization and Microsequencing. Springer-Verlag, Heidelberg, Germany, pp. 97–104.
Syv?oja,J., Suomensaari,S., Nishida,C., Goldsmith,J.S., Chui,G.S.J., Jain,S. and Linn,S. (1990) DNA polymerases , and : three distinct enzymes from HeLa cells. Proc. Natl Acad. Sci. USA, 87, 6664–6668.
Ishino,Y. and Cann,I.K.O. (1998) The euryarchaeotes, a subdomain of Archaea, survive on a single DNA polymerase: fact or farce? Genes Genet. Syst., 73, 323–336.
Hopfner,K.-P., Karcher,A., Craig,L., Woo,T.T., Carney,J.P. and Tainer,J.A. (2001) Structural biochemistry and interaction architecture of the DNA double-strand break repair Mre11 nuclease and Rad50-ATPase. Cell, 105, 473–485.
Moreau,S., Ferguson,J.R. and Symington,L.S. (1999) The nuclease activity of Mre11 is required for meiosis but not for mating type switching, end joining, or telomere maintenance. Mol. Cell. Biol., 19, 556–566.
Zhuo,S., Clemens,J.C., Stone,R.L. and Dixon,J.E. (1994) Mutational analysis of a Ser/Thr phophatase. J. Biol. Chem., 269, 26234–26238.
Mizuno,T., Yamagishi,K., Miyazawa,H. and Hanaoka,F. (1999) Molecular architecture of the mouse DNA polymerase alpha-primase complex. Mol. Cell. Biol., 19, 7886–7896.
Smith,R.W.P. and Nasheuer,H.-P. (2002) Control of complex formation of DNA polymerase alpha-primase and cell-free DNA replication by the C-terminal amino acids of the largest subunit p180. FEBS Lett., 527, 143–146.
Giot,L., Chanet,R., Simon,M., Facca,C. and Faye,G. (1997) Involvement of the yeast DNA polymerase in DNA repair in vivo. Genetics, 146, 1239–1251.
Dua,R., Levy,D.L. and Campbell,J.L. (1998) Role of the putative zinc finger domain of Saccharomyces cerevisiae DNA polymerase in DNA replication and the S/M checkpoint pathway. J. Biol. Chem., 273, 30046–30055.
Dua,R., Edwards,S., Levy,D.L. and Campbell,J.L. (2000) Subunit interactions within the Saccharomyces cerevisiae DNA polymerase (pol ) complex. Demonstration of a dimeric pol . J. Biol. Chem., 275, 28816–28825.
Shen,Y., Tang,X.-F. and Matsui,I. (2003) Subunit interaction and regulation of activity through terminal domains of the family D DNA polymerase from Pyrococcus horikoshii. J. Biol. Chem., 278, 21247–21257.
Motz,M., Kober,I., Girardot,C., Loeser,E., Bauer,U., Albers,M., Moeckel,G., Minch,E., Voss,H., Kilger,C. and Koegl,M. (2002) Elucidation of an archaeal replication protein network to generate enhanced PCR enzymes. J. Biol. Chem., 277, 16179–16188.
Shen,Y., Tang,X.-F., Yokoyama,H., Matsui,E. and Matsui,I. (2004) A 21-amino acid peptide from the cysteine cluster II of the family D DNA polymerase from Pyrococcus horikoshii stimulates its nuclease activity which is Mre11-like and prefers manganese ion as the cofactor. Nucleic Acids Res., 32, 158–168.
Furuse,M., Nagase,Y., Tsubouchi,H., Murakami-Murofushi,K., Shibata,T. and Ohta,K. (1998) Distinct roles of two separable in vitro activities of yeast Mre11 in mitotic and meiotic recombination. EMBO J., 17, 6412–6425.
Trujillo,K.M., Yuan,S.-S.F., Lee,E.Y.-H.P. and Sung,P. (1998) Nuclease activities in a complex of human recombination and DNA repair factors Rad50, Mre11 and p95. J. Biol. Chem., 273, 21447–21450.
Connelly,J.C., de Leau,E.S., Okely,E.A. and Leach,D.R.F. (1997) Overexpression, purification and characterization of the SbcCD protein from Eschericia coli. J. Biol. Chem., 272, 19819–19826.
Hopfner,K.-P., Karcher,A., Shin,D., Fairley,C., Tainer,J.A. and Carney,J.P. (2000) Mre11 and Rad50 from Pyrococcus furiosus: cloning and biochemical characterization reveal an evolutionary conserved multiprotein machinery. J. Bacteriol., 182, 6036–6041.
Glusker,J.P. (1991) Structural aspects of metal liganding to functional groups in proteins. Adv. Protein Chem., 42, 1–76.
Kunkel,T.A. (1988) Exonucleolytic proofreading. Cell, 53, 837–840.
Wilson,D.M,3rd (2003) Properties of and substrate determinants for the exonuclease activity of human apurinic endonuclease Ape1. J. Mol. Biol., 330, 1027–1037.
Miller,H. and Perrino,F.W. (1996) Kinetic mechanism of the 3' 5' proofreading exonuclease of DNA polymerase III. Analysis by steady state and pre-steady state methods. Biochemistry, 35, 12919–12925.
Leipe,D.D., Aravind,L. and Koonin,E.V. (1999) Did DNA replication evolve twice independently? Nucleic Acids Res., 27, 3389–3401.
Murzin,A.G. (1993) OB(oligonucleotide/oligosaccharide binding)-fold: common structural and functional solution for non-homologous sequences. EMBO J., 12, 861–867.
Arcus,V. (2002) OB-fold domains: a snapshot of the evolution of sequence, structure and function. Curr. Opin. Struct. Biol., 12, 794–801.
Agrawal,V. and Kishan,K.V.R. (2003) OB-fold: growing bigger with functional consistency. Curr. Protein Pept. Sci., 4, 195–206.
Theobald,D.L., Mitton-Fry,R.M. and Wuttke,D.S. (2003) Nucleic acid recognition by OB-fold proteins. Annu. Rev. Biophys Biomol. Struct., 32, 115–133.
Bateman,A., Birney,E., Cerruti,L., Durbin,R., Etwiller,L., Eddy,S.R., Griffiths-Jones,S., Howe,K.L., Marshall,M. and Sonnhammer,E.L.L. (2002) The Pfam protein families database. Nucleic Acids Res., 30, 276–280.
Evans,S.V. (1993) SETOR: hardware-lighted three-dimensional solid model representations of macromolecules. J. Mol. Graph., 11, 134–138, 127–128.(Maarit Jokela1, Anitta Eskelinen1, Helmu)