HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 32, 29846-29853, August 10, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/32/29846    most recent M103159200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brenkman, A. B. Articles by van der Vliet, P. C. Search for Related Content PubMed PubMed Citation Articles by Brenkman, A. B. Articles by van der Vliet, P. C. Social Bookmarking                      What's this?

The (I/Y)XGG Motif of Adenovirus DNA Polymerase Affects Template DNA Binding and the Transition from Initiation to Elongation^*

Arjan B. Brenkman, Marinus R. Heideman, Veronica Truniger^§, Margarita Salas^§, and Peter C. van der Vliet^¶

From the  University Medical Center, Department of Physiological Chemistry and Center for Biomedical Genetics, Utrecht, The Netherlands and ^§ Centro de Biología Molecular "Severo Ochoa," Universidad Autónoma, Canto Blanco, 28049 Madrid, Spain

Received for publication, April 10, 2001, and in revised form, May 30, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Adenovirus DNA polymerase (Ad pol) is a eukaryotic-type DNA polymerase involved in the catalysis of protein-primed initiation as well as DNA polymerization. The functional significance of the (I/Y)XGG motif, highly conserved among eukaryotic-type DNA polymerases, was analyzed in Ad pol by site-directed mutagenesis of four conserved amino acids. All mutant polymerases could bind primer-template DNA efficiently but were impaired in binding duplex DNA. Three mutant polymerases required higher nucleotide concentrations for effective polymerization and showed higher exonuclease activity on double-stranded DNA. These observations suggest a local destabilization of DNA substrate at the polymerase active site. In agreement with this, the mutant polymerases showed reduced initiation activity and increased K_m(app) for the initiating nucleotide, dCMP. Interestingly, one mutant polymerase, while capable of elongating on the primer-template DNA, failed to elongate after protein priming. Further investigation of this mutant polymerase showed that polymerization activity decreased after each polymerization step and ceased completely after formation of the precursor terminal protein-trinucleotide (pTP-CAT) initiation intermediate. Our results suggest that residues in the conserved motif (I/Y)XGG in Ad pol are involved in binding the template strand in the polymerase active site and play an important role in the transition from initiation to elongation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Adenoviruses contain a linear double-stranded genome of ~36 kilobases with two origins of replication located in the inverted terminal repeats. At each 5'-end of the adenovirus genome, a terminal protein (TP)¹ is covalently linked. Replication initiates via a protein-priming mechanism (1) involving the Ad pol and precursor terminal protein (pTP). Ad pol and pTP form a tight heterodimer of which the pTP acts as a primer and is covalently linked to the initiating nucleotide dCMP. Initiation of replication is catalyzed by Ad pol and can be stimulated by the two cellular transcription factors NFI and Oct-1, which function to recruit and position the pTP-pol complex on the origin of replication (Ref. 2 and references therein). Ad pol initiates replication opposite the fourth base of the template strand and synthesizes a pTP-CAT intermediate. For elongation to occur, this intermediate jumps back to be paired with template residues 1-3, after which pTP dissociates from Ad pol and elongation starts (3, 4). Elongation occurs via a strand displacement mechanism that requires the viral DNA-binding protein (reviewed in Refs. 5 and 6). Late in infection, pTP is cleaved by a virus-encoded protease into TP and the precursor part (reviewed in Refs. 5 and 6). The actual role of pTP processing is at present unclear. Initiation and elongation are performed by the same polymerase, but the two processes differ in sensitivity to inhibitors (7, 8). This suggests that a conformational change occurs upon transition from initiation to elongation, most likely after the formation of pTP-CAT. In agreement with this notion, kinetic studies revealed that the K_m for dCTP is lower for initiation than for elongation (9). In addition to its synthetic activities, Ad pol also possesses a distributive 3'-5'-exonuclease activity, shown to be involved in proofreading (10).

Many DNA polymerases have been characterized and were generally found to have a polymerase and a 3'-5'-exonuclease activity. Sequence comparisons of DNA polymerases from bacterial, viral, and cellular origin led to a classification into four groups, A, B (also known as -like), C, and D, based on amino acid similarities with Escherichia coli pol I, II, and III, and human DNA pol  (11-13). Based on their extent of similarity, six highly conserved motifs (I-VI), which were proposed to lie in the polymerase active site, were identified in human pol  (14). DNA polymerases containing these six motifs (see Fig. 1) were designated -like DNA polymerases (12). Further alignments showed that motifs I-III are conserved in all groups of DNA polymerases (12), while a seventh conserved motif was identified in the -like DNA polymerases (15, 16). Besides conserved motifs located in the C-terminal part of DNA polymerases, three sequence motifs (Exo I-III) were shown to form an evolutionary conserved 3'-5'-exonuclease site (17-19). Extensive biochemical analysis of a number of prokaryotic and eukaryotic DNA polymerases, such as the Klenow fragment of E. coli pol I (12), T4 (reviewed in Ref. 20), herpes simplex virus (21), 29 (reviewed in Ref. 22), and pol  (23), has shown that residues located in conserved motifs of these different DNA polymerases play similar roles in dNTP or DNA binding or in catalysis of polymerase or 3'-5'-exonuclease activity. Analysis of a number of DNA polymerases suggests that the polymerase active site is structurally and functionally conserved for both prokaryotic and eukaryotic DNA polymerases (24-28). They are all proposed to utilize an identical two-metal ion-catalyzed polymerase mechanism but differ extensively in many of their structural features (29). The crystal structure of phage RB69 DNA polymerase (24) can serve as prototype of the pol  family of DNA polymerases, since the recently solved crystal structures of Thermococcus gorgonarius DNA polymerase (30) and Thermococcus sp. 9 degrees N-7 (31) are topologically similar to this DNA polymerase.

Ad pol is an -like DNA polymerase belonging to the subclass of protein-priming DNA polymerases. Site-directed mutagenesis studies have identified motif I as a motif important for initiation and elongation activity of Ad pol (32). Furthermore, two putative zinc finger domains were identified (33), and linker mutagenesis studies have shown that multiple regions, including motifs IV and V, in Ad pol are essential for Ad DNA replication (34-36). Recently, a set of 22 alanine substitutions of conserved residues in the C-terminal part of Ad pol suggests an arrangement of conserved motifs in Ad pol similar to RB69 DNA polymerase (37).

An additional motif, YXG(G/A), located N-terminal of motif II (12, 18) of the polymerase active site, was shown to be highly conserved among -like DNA polymerases (38). Mutational analysis of this motif in 29 DNA polymerase, which starts replication by protein-priming, led to the proposal that it is involved in the binding stability of the DNA template at the polymerization active site (38). Additionally, it was shown to be important for the formation of a stable complex between TP and DNA polymerase, resulting in transition defects from TP priming to DNA priming during replication of 29 TP-DNA (39). A multiple alignment of the YXG(G/A) motif in eukaryotic-type DNA-dependent DNA polymerases has been shown previously by Truniger et al. (38). In eukaryotic-type DNA polymerases, the motif has the consensus sequence YXG(G/A), but for the subclass of protein-primed DNA polymerases, the motif could be restricted to the consensus YXGG. For these DNA polymerases, including Ad pol, the highly conserved tyrosine residue is often an isoleucine. This led us to define the motif as (I/Y)XGG.

Here, we report the detailed characterization of the (I/Y)XGG motif in Ad pol, which has been subjected to site-directed mutational analysis. We propose that the motif is involved in the stabilization of the template strand at the polymerase active site. During pTP-primed initiation, this indirectly affects the binding of the initiating nucleotide, as well as the transition of the initiation intermediate pTP-CAT from initiation to elongation, thereby leading to abortive replication. Based on the crystal structure of RB69, modeled with primer-template and dNTP, we propose a hydrophobic interaction between the conserved isoleucine and the ribose moiety of the nucleotide preceding the template base.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

DNA Templates and Substrates-- All oligonucleotides, unlabeled nucleotides, [-³²P]dNTPs (3000 Ci/mmol), and [-³²P]ATP (5000 Ci/mmol) were purchased from Amersham Pharmacia Biotech. T30 (5'-AATCCAAAATAAGGTATATTATTGATGATG) represents the first 30 nucleotides of the bottom strand of the adenovirus 5 genome, and T20 represents the first 20. D20 (5'-CATCATCAATAATATACCTT) is the complementary strand of T20. Labeling of D20 was performed with T4 polynucleotide kinase (Amersham Pharmacia Biotech) and [-³²P]ATP. D20 was used for the 3'-5'-exonuclease assay on ssDNA. For the polymerase/exonuclease coupled assay, gel retardation, and 3'-5'-exonuclease assay on dsDNA, 5'-labeled D20 was hybridized to T30 or to T20. The hybrid molecules D20/T30 and T20/D20 were obtained by boiling oligonucleotides in 60 mM Tris-HCl, pH 7.5, 200 mM NaCl, followed by slow cooling to room temperature. D20/T30 and D20/T20 were purified by 10% polyacrylamide-1× TBE gel electrophoresis. Ad 5 TP-DNA was isolated as described (40).

Site-directed Mutagenesis-- A full-length Ad pol cDNA encoding amino acids 1-1199 (provided by Henk G. Stunnenberg (41)) was cloned in the EcoRI and SphI sites of the pFastBac donor plasmid. Site-directed mutagenesis was performed using the QuickChange method from Stratagene. The oligonucleotides for the polymerase chain reaction mutagenesis were as follows: for R661A, 5'-ATGCTGGCGGCCACGTAATCG and 5'-CGATTACGTGGCCGCCAGCAT; for I664S, 5'-TCTTCCACCGCGGGAGCTGGCG and 5'- CGCCAGCTCCCGCGGTGGAAGA; for I664Y, 5'-TCTTCCACCGCGGTAGCTGGCG and 5'-CGCCAGCTACCGCGGTGGAAGA; for G666A/G667A, 5'-GTAGCATCTTGCAGCGCGGATGCTC and 5'-GAGCATCCGCGCTGCAAGATGCTAC, with changes marked boldface type.

The presence of the desired mutations was confirmed by complete sequencing of each mutant gene. The recombinant plasmids were transformed into DH10Bac competent cells, which contain the bacmid with a mini-attTn7 target site and a helper plasmid. The mini-Tn7 element on the pFastBac plasmid can transpose to the mini-attTn7 target site on the bacmid in the presence of transposition proteins provided by the helper plasmid. Colonies containing recombinant bacmids were identified by disruption of the lacZ gene. Bacmid DNA was isolated by means of a high molecular weight minipreparation. This DNA was then used to transfect insect cells with Lipofectin (Life Technologies) according to the manufacturer's manual. After 72 h of transfection, the recombinant baculoviruses were harvested and amplified for several rounds.

Expression and Purification of Ad DNA Polymerase Mutants-- Insect cells (Sf-9) were grown as monolayers on 167.5-cm² plates in SF900 II medium (Life Technologies) at 27 °C. Plates were infected with recombinant baculovirus expressing the wild-type or mutant Ad pol when ~80% confluence was reached. After 56 h of infection, cells were harvested and washed with ice-cold PBS. Cells were resuspended in a hypotonic lysis buffer containing 25 mM HEPES, pH 7.5, 10% glycerol, 5 mM KCl, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM DTT, 2 µg/ml aprotinin, and 1 µg/ml leupeptin and placed on ice. After 10 min, cells were disrupted by 20 strokes of a Dounce homogenizer (B pestle), and NaCl was added to a final concentration of 200 mM. The lysate was cleared by ultracentrifugation at 25,000 rpm in a SW28 rotor for 30 min at 4 °C.

For purification to near homogeneity, the lysate was loaded on a SP-Sepharose column, equilibrated with buffer A (25 mM HEPES, pH 7.5, 20% glycerol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM DTT) containing 200 mM NaCl. The column was washed extensively with buffer A, 200 mM NaCl and eluted with buffer A, 450 mM NaCl. Fractions were collected and analyzed on a 7.5% polyacrylamide, SDS gel followed by silver staining. Peak fractions were collected; dialyzed against buffer A, 100 mM NaCl; and loaded on a ssDNA-cellulose column (Sigma). After washing with buffer A, 150 mM NaCl, protein was eluted at 600 mM NaCl. Peak fractions were dialyzed against buffer B (25 mM HEPES, pH 8.0, 20% glycerol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM DTT) containing 100 mM NaCl and loaded onto a 1-ml Mono Q HR 5/5 column (Amersham Pharmacia Biotech). After washing with buffer B, 100 mM NaCl, protein was eluted in a gradient of buffer B, 100-500 mM NaCl. Peak fractions were collected, and the purity of the protein was estimated to be >95% by gel electrophoresis and Coomassie staining.

Proteins and Buffers-- pTP was a gift from Panagiotis N. Kanellopoulos. N-DNA-binding protein was purified as described (42). The buffer used for dilution of the replication proteins contained 25 mM HEPES, pH 7.5, 20% glycerol, 120 mM NaCl, and 1 mg/ml bovine serum albumin (BSA).

DNA Polymerase/Exonuclease Coupled Assay-- Partial duplex D20/T30 containing a stretch of 10 nucleotides protruding from the 5'-end was used as primer-template to study DNA-dependent DNA polymerization and 3'-5'-exonuclease activity. The reaction mixture (12.5 µl) contained 50 mM Tris-HCl, pH 7.5, 4% glycerol, 1 mM DTT, 1 mg/ml BSA, 1 mM MgCl₂, 0.05 ng of 5'-labeled D20/T30, 12.5 ng of wild-type or mutant DNA polymerase, and the indicated amounts of dNTPs. Reactions were stopped after 10 min at 37 °C by the addition of sequencing loading buffer (10 mM EDTA, 98% formamide, and 0.025% bromphenol blue). Samples were analyzed on 8 M urea-20% polyacrylamide gel electrophoresis followed by autoradiography. Polymerization or 3'-5'- exonuclease activity were detected as an increase or decrease in size, respectively, of the 5'-labeled D20 primer.

3'-5'-Exonuclease Assays-- Exonucleolytic breakdown of ssDNA and dsDNA was tested using 5'-labeled D20 and 5'-labeled D20/T30, respectively. The reaction mixture (25 µl) contained 50 mM Tris-HCl, pH 7.5, 4% glycerol, 1 mM DTT, 1 mg/ml BSA, 50 mM NaCl, 1 mM MgCl₂, 0.05 ng of ssDNA or dsDNA, and the reaction was started by the addition of 25 ng of wild-type polymerase or mutant polymerases. Incubation was at 37 °C, allowing conditions to be linear both in time and enzyme concentration. Reactions were stopped by the addition of sequencing loading buffer. After analysis by 8 M urea-20% polyacrylamide gel electrophoresis, the 3'-5'-exonuclease activity is measured as a decrease in size of the DNA by densitometry. From these data, the catalytic efficiency of the mutants (indicated in Table II) was calculated relative to wild-type Ad pol.

DNA Binding Assays-- Gel retardation was performed using 5'-labeled D20/T30 and 5'-labeled D20/T20. The binding reaction (20 µl) contained 25 mM HEPES, pH 7.5, 4% Ficoll, 1 mM EDTA, 55 mM NaCl, 4 mM DTT, 0.1 mg/ml BSA, 1 mM MgCl₂, 0.05 ng of either 5'-labeled D20/T30 or 5'-labeled D20/T20, and the indicated amounts of Ad pol or the corresponding mutants. After incubation for 5 min at 4 °C, samples were loaded and separated on a 10% polyacrylamide-1× TBE gel at 4 °C. Gels were dried, autoradiographed, and quantified using a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA).

Initiation and Partial Elongation of DNA Replication-- Initiation of replication was performed in a standard incubation mixture of 25 µl in the presence of 25 mM HEPES, pH 7.5, 50 mM NaCl, 1.5 mM MgCl₂, 1 mM DTT, 1 µg of BSA, 50 nM [-³²P]dCTP, and the indicated amounts of Ad pol and pTP. As template, either 0.6 µg of origin-containing T30 or 60 ng of TP-DNA were used. When TP-DNA was used, 250 ng of N-DNA-binding protein was added per reaction. Initiation coupled to partial/truncated elongation was performed under similar conditions as initiation in the presence of the indicated concentrations of dCTP, dATP, and dTTP. No dGTP was added in the reaction mixture. Reactions were performed at 37 °C for 45 min and were stopped by adding EDTA to a final concentration of 80 mM. The samples were precipitated with 20% trichloroacetic acid on ice. Precipitates were washed with 5% trichloroacetic acid, dissolved in sample buffer, analyzed on an SDS-7.5% polyacrylamide gel, and autoradiographed. Replication products were quantified by densitometric analysis following exposure on a PhosphorImager.

The K_m(app) for pTP deoxynucleotidylation was determined by performing initiation assays with [-³²P]dCTP with wild-type and mutant polymerases using increasing concentrations of unlabeled dCTP (1-1000 µM). The K_m(app) was calculated from three experiments.

Glycerol Gradient Sedimentation-- The standard incubation mixture (200 µl) for glycerol gradient analysis contained 2 µg of Ad pol, 1.2 µg of pTP, 25 mM HEPES, pH 7.5, 1 mM DTT, 1 mM MgCl₂, and NaCl to a final concentration of 55 mM. After incubation for 30 min on ice, the mixture was layered on top of a 4.8-ml linear 10-30% (v/v) glycerol gradient containing 25 mM HEPES, pH 7.5, 1 mM DTT, 1 mM EDTA, 0.5 M NaCl, and 100 µg of BSA as an internal control. Gradients were centrifuged for 24 h at 50,000 rpm in a SW50 rotor at 4 °C. A control gradient with 1.2 µg of pTP was run under similar conditions. Fractions were collected from the bottom of the tube and analyzed on an SDS-7.5% polyacrylamide gel. BSA was visualized by silver staining, and pTP, Ad pol, and the pTP-pol complex were visualized by immunoblotting using an anti-pol and anti-pTP-pol antiserum (43) Quantitation of the relative amounts of pTP and pol present in each fraction was carried out by densitometry.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In order to understand the role of the (I/Y)XGG motif in Ad pol, individual residues of this region were mutated (Fig. 1) as described under "Experimental Procedures." The isoleucine was changed into tyrosine (I664Y) as present in RB69, 29, and most other cellular, bacterial, and viral DNA polymerases (39) and also into serine (I664S) to study the effect of another nonconservative change. The two glycines are invariantly conserved among protein-primed DNA polymerases but the second glycine is often an alanine among bacterial, viral, and many cellular DNA polymerases (39). Both glycines were changed into alanines, giving mutant polymerase G666A/G667A. A positive charge preceding the (I/Y)XGG motif (Arg⁶⁶¹) appears to be specifically conserved among protein-primed and cellular DNA polymerases (38) and was changed into alanine (R661A). Construction of baculoviruses and expression and purification of the recombinant proteins was performed as described under "Experimental Procedures." During purification, all mutant polymerases behaved essentially as wild-type Ad pol.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Sequence conservation in Ad pol: relative location and alignment of the (I/Y)XGG motif. The relative location of the motifs conserved in eukaryotic-type DNA polymerases are indicated for Ad pol. Motif VI is lacking in Ad pol (14), and motif IV largely aligns with Exo II (19). In the lower panel, the motif (I/Y)XGG as defined in Ref. 38 is aligned for the DNA polymerases RB69, 29, and Ad pol (indicated as Ad-5), with the conserved amino acids in boldface type. An arginine, specifically conserved among protein-primed and cellular -like DNA polymerases, has been underlined. Mutant polymerases are shown at the bottom and are designated by the original amino acid (in single-letter notation), its position in Ad pol, and the replacing amino acid.

Mutations at the (I/Y)XGG Motif Alter the Wild-type Polymerase/Exonuclease Balance-- A polymerase/exonuclease coupled assay was performed to study the coordination of both degradative and polymerization activities with the mutant polymerases. The functional coupling between synthesis and degradation on primer-template D20/T30 was assayed as a function of the dNTP concentration. As shown in Fig. 2, 3'-5'-exonucleolytic digestion of the primer strand occurred in the absence of nucleotides. By the addition of increasing amounts of dNTPs, the equilibrium was shifted toward synthesis, exonucleolysis being competed by DNA polymerization. In the presence of 25 nM dNTPs, the wild-type Ad pol was able to extend D20/T30 until 27-28 nucleotides, and from 125 nM dNTPs full extension of the primer-template was accomplished (D30/T30). Whereas mutant polymerase R661A allowed full polymerization at approximately similar nucleotide concentrations as the wild-type enzyme (125 nM, Fig. 2), mutant polymerases I664S and I664Y required a 5-fold higher amount (Fig. 2 and Table I). On the other hand, mutant polymerase G666A/G667A required a 200-fold higher dNTP concentration compared with the wild-type enzyme for full polymerization (Fig. 2 and Table I). Furthermore, an increased exonuclease activity was observed for mutant polymerases I664S, I664Y, and G666A/G667A upon comparison of their degradation activities (Fig. 2, 0 nM lanes) with that of the wild-type polymerase, as can be seen by the higher intensity of the faster moving bands. The higher amount of dNTPs, required to fully elongate the primer-template for these three mutant polymerases, might be explained, at least partially, by their higher exonuclease activity. The 3'-5'-exonuclease activity was therefore determined on both ssDNA and dsDNA, and the results are quantified in Table I. Indeed, the exonuclease activity on primer-template DNA was increased for mutant polymerases I664S, I664Y, and G666A/G667A (Table I). Degradation of ssDNA by all mutant polymerases was slightly lower than that by the wild-type polymerase and proceeded in a distributive manner. An increased exonuclease activity on primer-template DNA might be the result of a lower DNA binding stability of the mutant polymerases in the polymerase active site.

View larger version (60K): [in this window] [in a new window]   Fig. 2.   DNA polymerase/exonuclease coupled assay. The assay was carried out using 5'-labeled D20/T30 primer-template, the indicated concentration of each dNTP, and 25 ng of wild-type or mutant Ad pol. The arrows indicate the position of the 20-mer (nonelongated primer) and the 30-mer (elongated primer).

DNA Binding of the Mutant DNA Polymerases Is Affected on Duplex DNA-- To examine the dsDNA binding capability of the different mutant DNA polymerases, gel retardation assays were performed with the primer-template molecule D20/T30 and duplex DNA molecule D20/T20. The formation of Ad pol-DNA complexes with the primer-template D20/T30 was similar for the wild-type and the mutant DNA polymerases (Fig. 3A). However, all mutant polymerases were affected in binding duplex DNA (Fig. 3B), although mutant polymerase R661A to a lesser extent than mutant polymerases I664S, I664Y, and G666A/G667A. As can be seen in Fig. 3, A and B, a second migrating band is visible for wild-type polymerase and on primer-template DNA for all mutant polymerases. A likely explanation for the presence of two migrating complexes is the existence of two forms of binding DNA by Ad pol (Ad pol monomeric and dimeric forms under conditions of low ionic strength (44)). These results are in agreement with the increased exonuclease activity found for three mutants on dsDNA (Table I) and indicate a decreased stability of the DNA in the polymerase active site.

View larger version (66K): [in this window] [in a new window]   Fig. 3.   DNA binding on duplex DNA is affected for all mutant polymerases. The gel retardation assay was performed using 5'-labeled dsDNA in the presence of the indicated amounts of wild-type or mutant Ad pol. Bands corresponding to free DNA and to polymerase-DNA complexes are indicated with an arrow. A, Ad pol binding to primer-template DNA D20/T30. B, Ad pol binding to duplex DNA D20/T20.

Initiation Activity Is Affected in All DNA Polymerase Mutants in the (I/Y)XGG Motif-- Since Ad pol uses pTP as primer in a template-dependent fashion during initiation, the formation of pTP-C was studied using T30 as template. As shown in Table II, the initiation activities of all mutant polymerases were severely affected with values of 14, 25, 17, and 11% compared with wild-type Ad pol for mutant polymerases I664S, I664Y, R661A, and G666A/G667A, respectively.

When the mutant polymerases were tested for initiation using TP-DNA (the adenovirus genome with covalently linked TP to each 5' DNA end), similar differences could be observed (Table II). Initiation performed without template or with an oligonucleotide of unrelated sequence did not result in any activity (data not shown), in agreement with previous data showing origin dependence (3, 45, 46). When initiation was performed with a 200-fold higher dCTP concentration, the initiation activity of the mutant polymerases showed an increase from 11-25 to 36-63% of the wild-type level (Table II), suggesting a reduced affinity for the initiating nucleotide, dCMP. Therefore, the apparent K_m (K_m(app)) for incorporation of dCTP was determined (Table II). Wild-type Ad pol showed a K_m(app) of 8.3 µM for the initiating nucleotide dCTP. Mutant polymerase R661A showed a K_m(app) value, which was only slightly increased compared with wild-type Ad pol. On the other hand, the K_m values of mutant polymerases I664S and I664Y were both 3-fold higher, while the K_m(app) of mutant polymerase G666A/G667A was shown to be 9 times higher than that of the wild-type polymerase (Table II).

The initiation impairment observed could be explained by an affected template strand binding of the mutated residues of the (I/Y)XGG motif. This could lead to an incorrect positioning of the templating nucleotide, explaining the increased K_m(app) shown for the initiating nucleotide. Another explanation for an increased K_m(app) for the initiating nucleotide could be a defective pTP/pol interaction. To discriminate between these possibilities, glycerol gradient analysis was performed as described under "Experimental Procedures" to study the interaction of pTP and the Ad pol mutants. Interaction with pTP was found to be stable, whereby all mutant polymerases behaved essentially as the wild-type Ad pol (Table II). It is therefore likely that the increased K_m(app) observed for the mutant polymerases is a result of destabilization of the template DNA strand, indirectly affecting dNTP binding. These results support the role of the (I/Y)XGG motif as a motif involved in binding template DNA.

Mutant Polymerase G666A/G667A Is Elongation-defective after Protein-primed Initiation-- Initiation on adenovirus DNA starts opposite the fourth nucleotide in the origin of replication with the formation of pTP-C, which extends to position 6, forming a pTP-CAT intermediate (3). The transition from initiation to elongation in adenovirus DNA replication is characterized by a jumping-back mechanism, which recovers the terminal three nucleotides, resulting in DNA primer-template elongation and dissociation of pTP (3, 4). To study elongation after protein priming, truncated replication reactions were performed in the presence of 50 nM dCTP and additionally 40 µM dATP and 40 µM dTTP but without dGTP. Under the conditions chosen (low dCTP), only part of the pTP-CAT intermediate is elongated until position 26 (the first C-residue in the template), thus providing an internal control for elongation efficiency. The pTP-26 product will migrate as a band of 90 kDa in SDS-polyacrylamide gels and is clearly distinguishable from the pTP-CAT intermediate (Fig. 4A). Wild-type Ad pol shows both the pTP-CAT intermediate and pTP-26 formation with a 5-fold higher intensity for pTP-26 than for pTP-CAT. Taking into account the five C residues in pTP-26, this indicates that equal amounts of pTP-CAT and pTP-26 were synthesized under these conditions, resulting in an elongation/initiation ratio of 1. Whereas the absolute levels of DNA synthesis were lower for mutant polymerases I664Y and R661A, the elongation to initiation ratios were similar to those of wild-type Ad pol. On the other hand, for mutant polymerase I664S, a pTP-26 product was visible only after long exposure, giving an elongation/initiation ratio of 0.01. For mutant polymerase G666A/G667A, no pTP-26 formation could be detected. When elongation was performed with increasing concentrations of dCTP, elongation activity could be partially restored for mutant polymerase I664S, increasing the elongation/initiation ratio to 0.2 (10 µM dCTP), while mutant polymerase G666A/G667A remained elongation-defective (Fig. 4B).

View larger version (43K): [in this window] [in a new window]   Fig. 4.   Protein-primed initiation and partial elongation on single-stranded origin containing DNA. Activity was assayed using 600 ng of T30 (the template DNA strand of the Ad 5 origin), 50 nM [-32P]dCTP, 90 ng of Ad pTP, and 100 ng of either wild-type or mutant Ad pol. A, initiation followed by partial elongation. In this assay, 40 µM each of dATP and dTTP were added to allow elongation up to 26 nucleotides. The positions of the pTP-CAT intermediate and the pTP-26 product are indicated. B, initiation followed by partial elongation as a function of the dNTP concentration. Partial elongation was allowed in the presence of 50 nM [-32P]dCTP, 40 µM each of dATP and dTTP, and increasing amounts of unlabeled dCTP.

Since mutant polymerase G666A/G667A showed no detectable elongation activity following initiation, we wondered whether initiation of mutant polymerase G666A/G667A resulted in the formation of pTP-CA and pTP-CAT intermediates. Therefore, initiation was assayed in the presence of labeled dATP or dTTP as shown in Fig. 5. To ascertain that the labeled nucleotide incorporated occupied the second or third position in pTP-CAT, respectively, 200 µM nonlabeled dCTP (in the case of [-³²P]dATP) or a 200 µM concentration of each dCTP and dATP (in the case of [-³²P]dTTP) was included in the reaction. When any of the three -³²P-labeled dNTPs were supplied separately as the only nucleotide, only dCMP could be directly linked to pTP (Ref. 3 and data not shown). Mutant polymerase G666A/G667A could incorporate three nucleotides to form the pTP-CAT intermediate. This result shows that mutant polymerase G666A/G667A is capable of dNTP incorporation using both the OH group of the serine (the priming amino acid of pTP) and the 3'-OH of nucleotides for polymerization, in agreement with the DNA-primed results (Fig. 2). However, DNA synthesis by mutant polymerase G666A/G667A stalls after pTP-CAT intermediate formation. Additionally, we observe a decrease in activity during each polymerization step from pTP-C to pTP-CAT, resulting in abortive replication for mutant polymerase G666A/G667A. These results suggest a defective translocation of the pTP-pol complex along the template DNA during the transition from initiation to elongation.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Mutant polymerase G666A/G667A is capable of pTP-CAT formation. Initiation was performed using 600 ng of T30, 90 ng of Ad pTP, and 100 ng of either wild-type or mutant Ad pol. Each reaction contained either 50 nM [-32P]dCTP marked as C; 200 µM dCTP and 50 nM [-32P]dATP marked as A; or 200 µM dCTP, 200 µM dATP, and 50 nM [-32P]dTTP marked as T. The percentage of initiation activity is based on the incorporation of each radiolabeled nucleotide relative to the wild-type polymerase and corrected for the elongation activity of mutant polymerases I664S and G666A/G667A.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The (I/Y)XGG motif (defined as YXG(G/A) in 29 DNA polymerase), highly conserved among eukaryotic-type DNA polymerases and located at the N-terminal site of the polymerase domain (38), was mutated in Ad pol to determine the function of these conserved residues in both DNA- and protein-primed reactions. Four mutant polymerases were obtained: R661A, I664S, I664Y, and G666A/G667A.

The (I/Y)XGG Motif in Ad Pol Stabilizes the Template Strand in the Polymerase Active Site-- Mutational analysis in the (I/Y)XGG motif of 29 DNA polymerase revealed an important role of this motif in the binding stability of the template strand at the polymerase active site (38). Different mutations in the same residue of this motif resulted in a pol/exo balance shifted either toward polymerization or toward exonucleolysis, depending on the stabilization of the template strand at a particular active site. During DNA-primed replication, three of the four mutant Ad pols described here showed a pol/exo balance shifted toward exonucleolysis (low pol/exo balance). Mutant polymerases I664S, I664Y, and G666A/G667A were shown to have a 2-3-fold increased exonuclease activity on dsDNA, while their activity on ssDNA was wild type-like. The increased exonuclease activity of these three mutant polymerases explains, at least partially, the higher dNTP concentrations required for effective polymerization on the same primer-template and indicates that DNA binding in the polymerase active site is affected. Indeed, when binding of duplex DNA was tested, the mutant polymerases were clearly affected in dsDNA binding in comparison with the wild-type Ad pol, indicating that a local destabilization of the template DNA at the polymerization active site exists in these mutant polymerases. However, no defective primer-template binding could be seen for any of the mutant polymerases. A possible explanation for this difference is that the mutant polymerases destabilize primer-template DNA only locally. Duplex DNA at the polymerase active site is held in position by numerous interactions of the fingers, palm, and thumb as has been shown in the crystal structure of Taq polymerase complexed with duplex DNA (25). However, in the presence of a 5'-template overhang, additional DNA contacts with the fingers are possible as shown in the crystal structure of bacteriophage T7 complexed with a primer-template (47). These additional contacts could further stabilize the primer-template at the polymerase active site, therefore affecting the DNA binding detectably only on duplex DNA. In the case of the 29 DNA polymerase mutants, defective retardation of dsDNA indicated a defective stabilization of the template strand at the polymerization active site and resulted in a low pol/exo balance (38). For mutant polymerase G666A/G667A the local destabilization of the primer-template at the polymerization active site might have an additional indirect effect on the positioning of the template strand. This could result in a template base that is incorrectly positioned for base pairing with the incoming nucleotide and thereby affect the affinity of this mutant polymerase for nucleotides during polymerization (200-fold higher nucleotide requirement).

The possible role of the (I/Y)XGG motif as a DNA binding motif is in agreement with the proposed localization of its residues (39) in the crystal structure of RB69 DNA polymerase modeled with DNA. The structure of RB69 DNA polymerase, one of the three DNA polymerases belonging to the -like DNA polymerases (family B) of which the crystal structure is known (24, 30, 31), can serve as a prototype for eukaryotic-type DNA polymerases. The RB69 polymerase active site modeled with a primer-template and a dNTP (crystallographic data from Protein Data Bank number 1WAH (39)) shows that the tyrosine of the (I/Y)XGG motif (Tyr³⁹¹, corresponding to Ile⁶⁶⁴ in Ad pol) interacts directly with the phosphate between the two nucleotides preceding the one acting as the template. Also, Gly³⁹³ and Ala³⁹⁴ are positioned close to the template strand. The motif is located more than 12 Å away from the dNTP binding site, indicating that a direct role of the (I/Y)XGG motif in dNTP binding is highly unlikely.

Decreased Template Binding Affects pTP-primed Initiation-- All four mutants described here were affected in pTP-primed initiation. The affected initiation activity and increase in the K_m(app) for the initiating nucleotide are likely a consequence of the local destabilization of the template strand at the polymerization active site, resulting indirectly in the incorrect positioning of the template strand for base pairing with the incoming nucleotide. This is in agreement with the results of the DNA binding experiments (Fig. 3) and with the position of the (I/Y)XGG motif in the RB69 structure, modeled with a primer-template and incoming dNTP (crystallographic data from Protein Data Bank number 1WAH (39)). A decrease in initiation activity for the (I/Y)XGG mutants of 29 DNA polymerase was also observed but was explained by an impaired interaction between TP and 29 pol (39). Since in 29 DNA polymerase protein-primed initiation is templated by the second nucleotide of the TP-DNA, the (I/Y)XGG motif could be located at the position of the TP during this reaction. Such an interaction defect is less likely for Ad pol during initiation, since it occurs opposite the fourth nucleotide, a situation where the motif is proposed to be in direct contact with the template strand (crystallographic data from Protein Data Bank number 1WAH (39)) and located rather distant from the protein primer. Indeed, we did not observe interaction differences between the wild-type and mutant polymerases and pTP as determined by glycerol gradient centrifugation. However, minor differences in interaction with pTP may not be detected in this assay. The increase in the K_m(app) for dCTP did not completely explain the decrease of the initiation activity in the case of mutant polymerase R661A. One possible explanation is that the interaction between pTP and Ad pol for mutant polymerase R661A was not fully functional, as has been shown to be the case for some 29 DNA polymerase mutants (48).

The (I/Y)XGG Motif Is Involved in the Transition from Initiation to Elongation-- Whereas the phenotypes of mutant DNA polymerases I664S and I664Y were mostly comparable, they clearly differed in their ability to elongate after protein priming. The elongation/initiation ratio of mutant polymerase I664Y was wild type-like, while mutant polymerase I664S showed a reduced elongation activity and was unable to restore the wild-type elongation to initiation ratio even at high dNTP concentrations. Mutant polymerase G666A/G667A had no detectable elongation activity even at high dNTP concentrations. However, both mutant polymerases, G666A/G667A and I664S, had been shown to be able to elongate a primer-template (Fig. 2) and were capable of pTP-CAT formation (Fig. 5). These results show that both mutant polymerases can use protein and DNA as a primer, and therefore the elongation defect of mutant polymerase G666A/G667A must lie in the transition from initiation to elongation. A decrease in the activity of mutant polymerases I664S and G666A/G667A during each additional polymerization step following initiation (C-CA-CAT) was observed, indicating a defective translocation of the pTP-pol complex along the template DNA. In the case of mutant polymerase G666A/G667A, this defective translocation led to abortive replication after the formation of the initiation intermediate pTP-CAT. The role of the (I/Y)XGG motif in stabilization of the template strand at the polymerase active site probably leads during initiation of protein-primed replication to a defect in the translocation of the template strand before, during, and after the jumping-back mechanism. A transition defect from initiation to elongation in TP-DNA has been described for three 29 mutant polymerases studied at the (I/Y)XGG motif (39). One of these mutated residues was Gly²²⁸, corresponding to Gly⁶⁶⁶ in Ad pol, which was changed into Ala. Their interaction defect with TP and/or DNA was proposed to cause a premature dissociation of TP, DNA polymerase, and DNA, resulting in the transition defect of the mutant polymerases. These mutant polymerases were not able to repolymerize short DNA fragments. Although a pTP-pol interaction defect was not found for the Ad pol mutants described here, they might not be able to efficiently bind the pTP-CAT intermediate. This would result in abortive replication. Glycine residues are often found in loops, as is Gly³⁹³, the first glycine of the (I/Y)XGG motif of RB69 polymerase. Thus, the glycine pair in Ad pol most likely functions as a structural element. The conservation of the glycine pair among many -like DNA polymerases suggests that it may play a critical role in creating the optimal environment for accommodation of the template strand. Recently, a study of Ad pol was carried out with crude lysates of 22 site-directed mutations to identify conserved residues involved in Ad pol function (37). In this study, mutant polymerases G667D and G666A/G667A from the (I/Y)XGG motif were tested for initiation activity, DNA binding, pTP-pol interaction, and polymerization activity on calf thymus DNA. Both mutant polymerases were mainly affected in initiation and dsDNA binding activity and were shown to interact with pTP, in agreement with our observations. In contrast, the initiation activity of mutant polymerase G666A/G667A was found to be much higher (50-75%) than in our study (11%). A likely explanation for this apparent discrepancy is that their assay conditions included nuclear extract that could contain cellular factors that might stimulate initiation. In addition, the presence of Mn²⁺ instead of Mg²⁺ used in the present study strongly reduces the specific nucleotide selection (3), which may lead to aberrant initiation products.

A Hydrophobic Contact That Positions the Template Strand-- In DNA polymerases of different Ad serotypes, the tyrosine in the (I/Y)XGG motif is an isoleucine. In contrast to tyrosine, isoleucine cannot hydrogen-bond with the phosphate between the two nucleotides preceding the one acting as a template as described for 29 pol (39). When the motif is aligned with the amino acid sequences of eukaryotic DNA-dependent DNA polymerases, the tyrosine is a phenylalanine in several bacterial and viral polymerases and an isoleucine in several TP-primed DNA polymerases, like Ad pol (38). This suggests that a hydrophobic residue at this position is important and might interact with the template DNA in the polymerase active site. Indeed, our results show an affected elongation for mutant polymerases I664S, whereas mutant polymerase I664Y has a wild type-like elongation/initiation ratio (Fig. 4A). Moreover, changing the tyrosine of the (I/Y)XGG motif of 29 pol into serine (Y226S) to keep the hydroxyl group resulted in a drastic phenotype with null polymerization and no dsDNA binding (38). Substitution into phenylalanine (Y226F), however, retained polymerization and dsDNA binding (38). We therefore propose that the isoleucine in the (I/Y)XGG motif of Ad pol interacts directly with the ribose moiety of the 3'-nucleotide preceding the templating nucleotide. This interaction might be important in the case of a tyrosine as well. Such a model is consistent with our results but also explains why the (I/Y)XGG motif in 29 pol controls the pol/exo balance immediately after the formation of the initiation product TP-dAMP (39). The TP-primed initiation of 29 pol starts opposite the second nucleotide of the template and uses a sliding-back mechanism to recover the terminal nucleotide (49). If the hydrogen bond between Tyr²²⁶ (the equivalent of Ile⁶⁶⁴ in Ad pol) and the phosphate between the two nucleotides preceding the template base were important, the pol/exo balance could only start at the formation of TP-(dAMP)₃. However, the pol/exo balance (which is due to the role of the motif for stably binding DNA at the polymerase active site (38)), starts at TP-(dAMP)₂ (39) when the hydrogen bond does not exist. This result is readily explained when the template is positioned correctly by a hydrophobic interaction with the nucleotide next to the template base. A similar hydrophobic interaction has been shown to be involved in discriminating between deoxynucleotides versus ribonucleotides in T7 DNA polymerase (47). Tyr⁵²⁶ together with Glu⁴⁸⁰ wedges the ribose moiety of the incoming nucleotide, thereby forming a hydrophobic pocket at the C2' position of the ribose that could exclude ribonucleotides from the polymerase active site (47).

In summary, our results support the conclusion drawn from the mutational analysis of 29 DNA polymerase that the (I/Y)XGG motif is involved in the coordination of synthesis and degradation due to its importance for the binding stability of the template DNA at the polymerization active site. Its role in DNA binding makes it additionally important for the transition steps from initiation to elongation. In addition, we propose that Ile⁶⁶⁴ directly interacts with the ribose moiety of the nucleotide next to the template nucleotide, which holds the template into position for correct initiation, jumping back, and subsequent elongation.

	ACKNOWLEDGEMENTS

We thank Rob de Jong, Kevin Augustijn, and Bas van Breukelen for critical reading of the manuscript and Rob de Jong and Miguel de Vega for stimulating discussions.

	FOOTNOTES

^* This work was supported in part by the Netherlands Organization for scientific research (to P. C. v. d. V.), by National Institutes of Health Grant 2R01 GM27242-21 and Dirección General de Investigación Científica y Técnica grant PB98-0645 (to M. S.), by an institutional grant from the Fundación Ramón Areces (to the Centro de Biología Molecular "Severo Ochoa"), and by European Union Contract FMRX-CT97-0125 (to P. C. v. d. V. and M. S.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Universiteitsweg 100, P.O. Box 85060, 3508 AB, Utrecht, The Netherlands. Tel.: 31-302538989; Fax: 31-302539035; E-mail: p.c.vandervliet@med.uu.nl.

Published, JBC Papers in Press, June 4, 2001, DOI 10.1074/jbc.M103159200

	ABBREVIATIONS

The abbreviations used are: TP, terminal protein; Ad, adenovirus; pol, polymerase; exo, exonuclease; pTP, precursor terminal protein; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; DTT, dithiothreitol; BSA, bovine serum albumin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				E. Longas, M. de Vega, J. M. Lazaro, and M. Salas Functional characterization of highly processive protein-primed DNA polymerases from phages Nf and GA-1, endowed with a potent strand displacement capacity Nucleic Acids Res., November 6, 2006; 34(20): 6051 - 6063. [Full Text] [PDF]

				M. E. Mysiak, C. Wyman, P. E. Holthuizen, and P. C. van der Vliet NFI and Oct-1 bend the Ad5 origin in the same direction leading to optimal DNA replication Nucleic Acids Res., December 1, 2004; 32(21): 6218 - 6225. [Abstract] [Full Text] [PDF]

				M. E. Mysiak, P. E. Holthuizen, and P. C. van der Vliet The adenovirus priming protein pTP contributes to the kinetics of initiation of DNA replication Nucleic Acids Res., July 25, 2004; 32(13): 3913 - 3920. [Abstract] [Full Text] [PDF]

				M. E. Mysiak, M. H. Bleijenberg, C. Wyman, P. E. Holthuizen, and P. C. van der Vliet Bending of Adenovirus Origin DNA by Nuclear Factor I as Shown by Scanning Force Microscopy Is Required for Optimal DNA Replication J. Virol., February 15, 2004; 78(4): 1928 - 1935. [Abstract] [Full Text] [PDF]

				B. van Breukelen, A. B. Brenkman, P. E. Holthuizen, and P. C. van der Vliet Adenovirus Type 5 DNA Binding Protein Stimulates Binding of DNA Polymerase to the Replication Origin J. Virol., December 20, 2002; 77(2): 915 - 922. [Abstract] [Full Text] [PDF]

				A. B. Brenkman, E. C. Breure, and P. C. van der Vliet Molecular Architecture of Adenovirus DNA Polymerase and Location of the Protein Primer J. Virol., July 17, 2002; 76(16): 8200 - 8207. [Abstract] [Full Text] [PDF]

				X. Wang and J. Hu Distinct Requirement for Two Stages of Protein-Primed Initiation of Reverse Transcription in Hepadnaviruses J. Virol., May 13, 2002; 76(12): 5857 - 5865. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/32/29846    most recent M103159200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Brenkman, A. B. Articles by van der Vliet, P. C. Search for Related Content PubMed PubMed Citation Articles by Brenkman, A. B. Articles by van der Vliet, P. C. Social Bookmarking                      What's this?