The Roles of Tyr391 and Tyr619 in RB69 DNA Polymerase Replication Fidelity

Abstract

In the family-B DNA polymerase of bacteriophage RB69, the conserved aromatic palm-subdomain residues Tyr391 and Tyr619 interact with the last primer-template base pair. Tyr619 interacts via a water-mediated hydrogen bond with the phosphate of the terminal primer nucleotide. The main-chain amide of Tyr391 interacts with the corresponding template nucleotide. A hydrogen bond has been postulated between Tyr391 and the hydroxyl group of Tyr567, a residue that plays a key role in base discrimination. This hydrogen bond may be crucial for forcing an infrequent Tyr567 rotamer conformation and, when the bond is removed, may influence fidelity. We investigated the roles of these residues in replication fidelity in vivo employing phage T4 rII reversion assays and an rI forward assay. Tyr391 was replaced by Phe, Met and Ala, and Tyr619 by Phe. The Y391A mutant, reported previously to decrease polymerase affinity for incoming nucleotides, was unable to support DNA replication in vivo, so we used an in vitro fidelity assay. Tyr391F/M replacements affect fidelity only slightly, implying that the bond with Tyr567 is not essential for fidelity. The Y391A enzyme has no mutator phenotype in vitro. The Y619F mutant displays a complex profile of impacts on fidelity but has almost the same mutational spectrum as the parental enzyme. The Y619F mutant displays reduced DNA binding, processivity, and exonuclease activity on ssDNA and dsDNA substrates. The Y619F substitution would disrupt the hydrogen-bond network at the primer terminus and may affect the alignment of the 3′ primer terminus at the polymerase active site, slowing chemistry and overall DNA synthesis.

Introduction

Replicative DNA polymerases bear the primary responsibility for accurate DNA replication. The DNA polymerase of phage RB69, gp43, is encoded by gene 43, a homologue of T4 DNA polymerase, which belongs to the B family of DNA polymerases and bears marked sequence similarities with eukaryotic replicative polymerases such as α, δ and ε. gp43 has both polymerase and exonuclease domains on one polypeptide. This protein is responsible for the high fidelity of RB69 replication, introducing about one mutation for every 2 × 10⁸ nucleotides incorporated.^1,2 This high fidelity is achieved by selection of the correct nucleotide in the polymerase (Pol) domain (error rate 10⁻⁵) and subsequent proofreading in the exonuclease (Exo) domain (error rate 2 × 10⁻³); the host DNA mismatch repair system does not operate on T4 DNA and presumable also not on RB69 DNA, and T4 seems to lack a general mismatch repair system of its own.^3,4 A tight nucleotide-binding pocket (Pol pocket) formed by the movement of the fingers subdomain from an open to a closed position is responsible for discriminating between correct and wrong nucleotides.⁵ Proofreading by the associated 3′-5′ exonuclease activity removes most misincorporated nucleotides.

Several crystal structures for RB69 DNA polymerase provide opportunities to study the contribution of individual amino acids to fidelity. Several Pol-pocket mutants were investigated for their abilities to discriminate correct versus incorrect dNTPs.^2,6,7 We reported previously that the RB69 Pol-pocket Y567A mutant exhibits strongly reduced base selectivity and enhanced mutagenicity.² Another Pol-pocket residue, Leu561, was also found to play a role in discriminating against mismatches.⁷

In addition to residues that form the Pol binding pocket, structural studies of replicative polymerases in the RT, A and B families indicate that polymerases also monitor correct base-pairing geometry through minor-groove interactions up to 4–5 base pairs upstream of the active site.^8–11 The primer-template DNA maintains a B-form conformation in the RB69 Pol domain, in contrast to the A-form DNA observed in A-family polymerases.^9,12–14 The B-form DNA is stabilized by a conserved KKRY motif (residues 705-708 in RB69 gp43). Tyr708 from this motif makes a hydrogen bond to the phosphodiester at the primer 3′ terminus, while Lys705 and Arg707 interact with the template-strand phosphates.¹⁵ These residues constrain the primer and template strands and orient the primer terminus in the Pol site.¹⁵ Gp43 main and side chains interact via hydrogen bonds with the DNA phosphodiester backbone, but these residues are not conserved.

Here we focus on two residues, Tyr391 and Tyr619, that are not part of the nucleotide binding pocket but make contacts with the DNA upstream of the Pol pocket (Figure 1). In the closed ternary complex (gp43, DNA primer-template, and incoming nucleotide), Tyr619 interacts (via a water-mediated hydrogen bond) with the phosphate of the terminal primer nucleotide; Tyr619 is conserved in most of the B-family polymerases.^5.15 The main-chain amide of Tyr391 interacts with the phosphate 5′ to the template base of the last template-primer base pair. Tyr391 may also form a hydrogen bond with the hydroxyl group of Tyr567; both Tyr391 and Thr587 were predicted to be crucial for restricting the infrequent Tyr567 rotamer conformation.⁶ Based on the crystal structure, this rotamer conformation seems to provide a steric gate for checking Watson-Crick base pairing.

Figure 1.

(a) Cartoon of molecular interactions involving polymerase residues Tyr391 and Tyr619. P = backbone phosphate, PB = primer base, IB = incoming base, TB = template base, and arrows indicate amino-acid/DNA or amino-acid/amino-acid interactions (see text). (b) A view of the minor groove edge of a nascent base pair (TTP—dA) binding pocket illustrating several key hydrogen-bonding interactions (black dotted lines) (PDB ID code 1IG9). Side-chain carbon atoms of the fingers and palm polymerase subdomains are colored green and purple, respectively. The tyrosine side chains altered in this study (Y391 and Y619) are shown in a stick representation, whereas surrounding residues are illustrated in a wire representation. The primer and template strands are colored yellow and orange, respectively, and the 5′-end of the template strand is indicated. This image was created by Bill Beard in Chimera.³⁰

The aim of this work was to examine the role of these two residues in the fidelity of RB69 DNA polymerase. Tyr619 was replaced by phenylalanine, and Tyr391 was replaced by alanine, methionine, and phenylalanine, and the fidelity of the modified polymerases was measured in several mutation assays. Both mutator and antimutator consequences were observed.

Results

Substitutions at Tyr391

In the crystal structure of RB69 gp43 in the closed conformation, the conserved Tyr391 contacts the phosphate between the (n – 2) and (n – 1) template nucleotides, and Tyr391 and Thr587 are involved in stabilizing an infrequent conformation of Tyr567 that plays a crucial role in base selection.⁶ The replacement Y391F had no effect on pre-steady-state kinetic parameters for incorporation of complementary or noncomplementary dNMPs. The replacement Y391A increased the K_d for incoming complementary dNTPs about 14-fold, hardly changed k_pol, and decreased discrimination against incoming dGTP opposite template G about 10-fold.⁶ To elucidate the role of Y391 in replication fidelity, we first constructed the replacements Y391A, Y391M and Y391F in both Exo⁺ and Exo⁻ backgrounds and then asked whether these mutants can support the growth of T4 in vivo. T4 43am phages totally deficient in polymerase were used to infect cells carrying a plasmid expressing one or another of the RB69 gene-43 constructs and spot tests were used to assess growth (Table 1). The Y391M and Y391F replacements supported substantial growth in either Exo background, but the Y391A construct was inviable. However, cells expressing Y391A supported the growth of T4 43⁺, indicating that the RB69 43Y391A mutation is recessive. We also measured T4 DNA synthesis in infected cells at a time when host DNA synthesis had ceased completely (Table 1). With the Y391A mutant in either Exo configuration, DNA synthesis is reduced by 90 %, a level (and perhaps quality) of DNA synthesis that seems to be insufficient to support T4 replication. In contrast, DNA synthesis was robust with the Y391F construct and moderate in the Y391M construct.

Table 1.

Relative DNA synthesis

Polymerase	Growth on T4 43amam	Growth on T4 43+	Relative DNA synthesis
Pol+ Exo+	+	+	1.0
Pol+ Exo−	+	+	0.9
PolY391M Exo+	+	+	0.3
PolY391M Exo−	+	+	0.3
PolY391F Exo+	+	+	0.7
PolY391F Exo−	+	+	0.8
PolY391A Exo+	−	+	0.1
PolY391A Exo−	−	+	0.1
PolY619F Exo+	+	+	0.3
PolY619F Exo−	+	+	0.3
PolY619A Exo+	+	+	0.24

Phage growth was estimated by spot tests. relative DNA synthesis was normalized to [³H]thymidine incorporation 27–37 min after infection with T4 43am. The value “1” for Pol⁺ Exo⁺ represents 1.5 × 10⁵ dpm ³H per 10⁷ infected cells.

rII reversion tests were performed using the two viable constructs Y391F and Y391M in either an Exo⁺ background (Table 2) or an Exo⁻ background (Table 3). The Y391F Exo⁺ enzyme is a weak mutator for both ±1 frameshift mutations and base-substitution mutations, but Y391F Exo⁻ displays only slightly enhanced +1 frameshift mutation, a small antimutator effect at the G:C site, and no significant impact at the other two sites. The Exo⁻ results tend to agree with the kinetic data showing no differences between the parental and mutant enzymes.⁶ The Y391M polymerase has at most a marginal impact in the reversion tests in either Exo background.

Table 2.

Mutator activities of Exo⁺ RB69 Pol^Y619F, Pol^Y391F and Pol^Y391M in reversion tests

		Reversion rate
rII tester mutation	Mutations scored	Pol+	PolY391F	PolY391M	PolY619F
131	+1	9.0	33.7 (3.7)	2.6 (0.3)	0.96 (0.1)
UV232	−1	1.0	3.3 (3.3)	2.8 (2.8)	3.0 (3.0)
UV256	G·C→A·T	6.4	16.4 (2.6)	14.7 (2.3)	8.9 (1.4)
UV375	A·T→ any	1.6	3.5 (2.2)	0.9 (0.6)	0.6 (0.4)

Reversion rates are per 10⁸ and are medians of 17 stocks for Pol⁺, 12–18 stocks for Pol^Y619F, and 7 stocks for Pol^Y391F and Pol^Y391M. Values in parentheses are relative to the Pol⁺ rate.

Table 3.

Mutator activities of Exo⁻ RB69 Pol^Y619F, Pol^Y391F and Pol^Y391M in reversion tests

		Reversion rate
rII tester mutation	Mutations scored	Pol+	PolY391F	PolY391M	PolY619F
131	+1	2700	7550 (2.8)	2480 (0.9)	1000 (0.4)
UV232	−1	1580	1170 (0.7)	673 (0.4)	177 (0.1)
UV256	G·C→A·T	8750	2280 (0.3)	5370 (0.6)	1430 (0.2)
UV375	A·T→ any	1360	1340 (1.0)	990 (0.7)	277 (0.2)

Reversion rates are per 10⁸ and are medians of 17 stocks for Pol⁺, 12–18 stocks for Pol^Y619F, and 7 stocks for Pol^Y391F and Pol^Y391M. Values in parentheses are relative to the Pol⁺ rate.

Because the Y391A mutant was inviable, the enzyme was purified and examined in the M13mp2 gap-filling assay in vitro (Table 4). This polymerase could fill a 400-nt gap. There is no discernable mutator effect in the Exo⁺ background, although the high intrinsic background in the lacZα system could conceal a substantial mutator factor. The modest mutator effect in the Exo⁻ background is of marginal reliability in this test. Thus, the inability of the Y391A polymerases to support growth in vivo probably reflects an inability to synthesize DNA of good quality (for instance, of high molecular weight) rather than a strong mutator effect leading to mutational meltdown as seen previously with another construct.²

Table 4.

Pol^Y391A forward-mutation frequencies in vitro

Polymerase	Total plaques	Mutant plaques	Correction factor	lacZα mutation frequency × 104
Pol+ Exo+	54411	52	0.88	8
Pol+ Exo−	91799	276	0.95	30
PolY391A Exo+	69213	71	1	10
PolY391A Exo−	30848	120	1	40

Entries for Pol⁺ Exo⁺ and Pol⁺ Exo⁻ RB69 DNA polymerases are from a previous report.¹⁹ Correction factors reflect occasional light-blue mutants with no sequence change in the 293-bp lacZα sequence; we did not sequence mutants from Pol^Y391A, so the correction factors were assumed to be 1.

Substitutions at Tyr619

In the crystal structure of RB69 gp43 in the closed conformation, the conserved Tyr619 contacts DNA from the primer side via a water-mediated hydrogen bond with the phosphate of the last base pair, and also contacts the conserved residue Asp621. Tyr619 lies in a cluster of hydrophobic amino acids. In most other B-family polymerases this Tyr is conserved, although T4 gp43 has an alanine at the corresponding position. Tyr619 also forms a hydrogen bond to the main-chain nitrogen of Lys706, part of the highly conserved KKRY motif. Is Tyr619 important for fidelity? To answer this question, we constructed Y619F. In either Exo background, the Y619F enzyme supported T4 replication in vivo and synthesized DNA at a modest 30% rate (Table 1). In rII reversion assays, the Y619F mutant polymerase displays a somewhat complex pattern (Tables 2 and 3). In the single-base addition test it displays an antimutator phenotype, particularly in the Exo⁺ background. In the single-base deletion test it displays a weak mutator effect in the Exo⁺ background but a substantial antimutator effect in the Exo⁻ background. It has no significant impact on base-pair substitutions in the Exo⁺ background, but is a moderate antimutator in the Exo⁻ background.

Specific mutabilities of different base pairs vary greatly, potentially confounding the interpretation of many reversion tests and kinetic analyses. In order to explore possible context-dependent effects on fidelity in Tyr619 constructs, we turned to the T4 rI in vivo forward-mutation system. In contrast to the lacZα system, the rI system does not suffer an inflated background mutation frequency. As shown in Table 5, Pol^Y619F Exo⁺ displays a 5.4-fold mutator effect averaged over many sites and kinds of mutations, while Pol^Y619F Exo⁻ displays a 5.7-fold antimutator effect compared to Pol⁺ Exo⁻. Thus, the complexity of fidelity effects with the Y619F mutant are reproduced in the rI assay.

Table 5.

Pol^Y619F polymerase forward-mutation rates in vivo

Polymerase	Number of stocks	μrI × 105	Rel. μrI
Pol+ Exo+	19	0.43	1
Pol+ Exo−	7	220	510
PolY619F Exo+	12	2.3	5.4
PolY619F Exo−	12	38	89

μ_rI entries are mutation rates for the T4 rI gene, for which median values are posted.

DNA sequencing provides a finer resolution of the rI rates (Table 6). The overall conclusion from Table 6 is that the mutator effects of Y619F in an Exo+ background are quite smoothly distributed over most the classes of mutations, albeit with some preference for transversions. The antimutator effects of Y619F in an Exo⁻ background are similarly smoothly distributed, again with a greater average impact on transversions. Even the indels and the complex mutations are similarly impacted; most of the latter are GCG → CTA arising by a templating mechanism involving template switching.¹⁶ This overall uniformity is also obvious from the mutational spectra (Figure 2), the only difference being a strong hotspot at 247-248, for G → A transitions at 247 and for C → T transitions at 248 in the Pol^Y619F Exo⁻ spectrum, which is almost absent in the Pol⁺ Exo⁻ spectrum. Also, in the reversion assay, Pol^Y619F Exo⁺ has a lower +1 mutation rate at a run of 5 A:T base pairs, but not in the forward-mutation assay.

Table 6.

Kinds and rates of rI mutations generated by Pol^Y619F

	Pol+ Exo+		PolY619F Exo+		Pol+ Exo−		PolY619F Exo−
Change	No.	μ × 107	No.	μ × 107	No.	μ × 107	No.	μ × 107
Mutants	102		104		96		97
Mutations	103	42.8	105	231	103	22000	98	3820
G→A	17	7.1	11	24	19	4050	23	900
C →T	5	2.1	3	6.6	9	1900	14	550
A→G	2	0.8	0	≤2	5	1050	2	78
T→C	5	2.1	1	2.2	14	3000	2	78
Transitions	29	12.1	15	33	47	10000	41	1610
A→C	1	0.4	1	2.2	0	≤200	1	39
T→G	1	0.4	0	≤2	2	430	0	≤40
A→T	0	≤0.4	1	2.2	2	430	2	78
T→A	0	≤0.4	1	2.2	7	1500	0	≤40
G→T	13	5.4	23	51	10	2100	5	195
C→A	21	8.7	29	64	7	1500	6	230
Transversions	36	15.0	55	121	28	6000	14	550
+1	5	2.1	4	8.8	22	4700	32	1250
−1	5	2.1	5	11.0	5	1100	9	350
≥±2	7	2.9	8	18	0	≤200	1	39
Indels	17	7.1	17	37	27	5800	42	1640
Complex	21	8.7	18	40	1	200	1	39

The Pol⁺ Exo⁺ and Pol⁺ Exo⁻ entries are based on a previous collection² supplemented with more recently collected mutants.

Figure 2.

Pol^Y619F mutational spectra. The sequence is the wild-type rI strand complementary to the coding strand with every tenth base in bold face. Capital letters indicate base substitutions. The deletion of a single base is indicated by − and the addition of a single base by +. The deletion or addition of two or more bases is indicated by − or +, respectively, before the underlined capital letters. The addition of a novel base is indicated by + before the base, both underlined, and overlying the two mutationally separated bases. The complex mutation GCG →CTA at 146–148 appeared 17 times in the Pol⁺ Exo⁺ spectrum and 15 times in the Pol^Y619F Exo⁺ spectrum. (a) Mutational spectra for Pol^Y619F Exo⁺ (above) and Pol⁺ Exo⁺ (below). (b) Mutational spectra for Pol^Y619F Exo⁻ (above) and Pol⁺ Exo⁻ (below).

To better understand the properties of the Y619 enzyme, we purified the parental and mutant polymerases and investigated their DNA binding affinities, exonuclease activities, and processivities.

The capacity of the Y619F polymerase to bind the primer terminus was assessed in a gel-retardation assay. As shown in Figure 3, the Y619F protein shows strongly reduced binding. The reactions contained the same amount of substrate (primer-template hybrid DNA 20/35 mer) and increasing concentrations of enzymes. For both the Pol⁺ Exo⁻ and Pol^Y619F Exo⁻enzymes, binding increased proportionally to enzyme concentration. At the highest enzyme concentration, almost 90% of the DNA substrate was bound by the Pol⁺ Exo⁻ polymerase but only 20% by the Pol^Y619F Exo⁻ polymerase. The Pol^Y619F Exo⁻ polymerase is ~5-fold weaker in DNA binding.

Figure 3.

Polymerase binding to DNA. Affinities of Exo⁻ Pol⁺ and Pol^Y619F polymerases were determined by a mobility-shift assay with a double-stranded 20-mer/35-mer primer-template as described in the Methods. Primer-template (50 nM) was mixed with 2.5, 5, 10, 25, 50, 100 and 200 nM of the indicated protein. Quantitative analysis of autoradiograms was performed using the Storm Phosphoimager. The amount of bound substrate was quantified and plotted as a function of enzyme concentration. Data are averages from at least 3 experiments. Two shifted products visible at the highest concentration of Y619F polymerase may correspond to one or more RB69 molecules bound per oligonucleotide substrate.³¹

In addition to facilitating the velocity of genome replication, polymerase processivity, the number of nt incorporated per polymerase association/dissociation cycle with the template-primer, may relate to frameshift fidelity. We measured the processivities of Pol⁺ Exo⁻ and Pol^Y619F Exo⁻ polymerases on a primed M13mp2 lacZα template. The Pol^Y619F Exo⁻ polymerase is less processive than the Pol⁺ Exo⁻ enzyme, adding only 20 nt per binding event while the Pol⁺ Exo⁻ polymerase can extend this particular primer up to 85 nt (Figure 4). In keeping with the observation that the T4 and RB69 polymerases have difficulty overcoming even weak secondary structures in the template strand in the absence of accessory proteins,¹⁷ the Pol⁺ Exo⁻ enzyme terminates at a template position where a weak secondary hairpin structure probably forms. Mutant polymerase dissociates more often from the primer-template than does parental enzyme.

Figure 4.

Polymerase processivities. Processivity reactions were analyzed on 12% denaturing polyacrylamide gel. Products of 5-, 10- and 15-min reactions are shown for Exo⁻Pol⁺ (lanes 1–3) and Pol^Y619F (lanes 4–6) polymerases. The molar ratios of primer-template to enzyme are indicated. A DNA sequencing ladder is shown in the last four lanes. The numbers on the left are template nucleotides. The Pol⁺ polymerase terminates synthesis at the palindrome 5′-AATTGTGAGCGGATATAACAATT-3′ where a weak secondary structure may form.

Proofreading by the 3′-exonuclease strongly enhances the overall fidelity of the T4 and RB69 DNA polymerases. The Pol and Exo domains interact, so that mutations at the Pol domain often affect Exo activity¹⁸ or proofreading efficiency.² Exonuclease assays revealed that the Pol^Y619F Exo⁺ enzyme has reduced Exo activities on both ssDNA and dsDNA substrates (Figure 5). Thus, the Y619F substitution reduces not only polymerase but also exonuclease activity. In reversion tests (Tables 2 and 3), mutation rates in Pol^Y619F relative to Pol⁺ are usually greater in Exo⁺ than in Exo⁻ backgrounds, which suggests that the Y619F substitution mildly impairs partitioning of the DNA primer terminus to the Exo site.

Figure 5.

Exonuclease assays. Reactions were incubated at 37 °C for 10, 20, 40, 80, 180, and 300 sec. Degradation of the labeled DNA was analyzed by electrophoresis in 7 M urea/15% polyacrylamide gels followed by autoradiography. (a) ssDNA exonuclease assay using a ³²P-labelled 20-mer ssDNA substrate. (b) dsDNA exonuclease assay run using a ³²P-labelled 20-mer/35-mer substrate. The nM of substrate remaining as a function of time is plotted for each exonuclease assay.

Discussion

Role of Tyr391

It has been suggested that the well conserved Tyr391 residue helps to stabilize a Tyr567 rotamer conformation,⁶ which in turn plays a major role in base discrimination.^2,19 Thus, substituting Tyr391 with Ala, Met or Phe, which would disrupt its putative hydrogen bonding with Tyr567, should also impact fidelity and probably polymerizing activity. Both the Phe and Met replacements do affect both fidelity and activity, but by only small factors (Tables 1–3). The alanine replacement produces a polymerase with severely reduced capacity in vivo but which can synthesize DNA in vitro (Tables 1 and 4), where it displays no significant mutator activity (Table 4).

The structure of a ternary complex of RB69 gp43 shows that Tyr391 is part of the “floor” of the nascent base-pair platform in the palm subdomain.¹⁵ The Tyr391 aromatic ring forms van der Waals contacts with the ribose ring of the template (n – 1) nucleotide. The Tyr391 hydroxyl group forms a putative hydrogen bond with the hydroxyl group of Tyr567 in the fingers subdomain, and Tyr567 may also be hydrogen bonded with Thr587. Another tyrosine residue in close proximity to Tyr391 is Tyr416, which interacts with the ribose ring of the incoming nucleotide. The role of this region in G·C base selectivity has been studied in detail,⁶ providing pre-steady-state parameters for the incorporation of the correct and incorrect nucleotide by Y391A/F and Y567A/F replacements of the Exo⁻ polymerase. The results indicate that the hydrogen bond between Tyr391 and Tyr567 plays a role in stabilizing Tyr567 in an infrequent rotamer conformation, which is essential for forming the nucleotide binding pocket and which provides selectivity among incoming nucleotides.⁶

In agreement with kinetic studies showing a substantial increase in the K_d of Y391A polymerase towards correct nucleotides,⁶ this mutant cannot sustain DNA replication in vivo (Table 1), where the required high dNTP concentrations are not available. The most likely reason for the decreased affinity of the Y391A enzyme for incoming dNTPs is a diminished closed configuration replacing the van der Waals contact between the Tyr side chain and the ribose ring of the (n – 1) template nucleotide, which could allow movement of template DNA resulting in misalignment of an incoming dNTP base with the corresponding template base. The conformation of amino-acid side chains in the binding pocket would not be much affected, in accordance with the observation that the Y391A k_pol is not changed.

The Y391M mutant can replicate in vivo even though its rate of DNA synthesis is 70% lower than the wild type (Table 1). The fact that mutation to the larger Met residue is less damaging than mutation to Ala agrees with the mechanism proposed above. Replacing Tyr391 with a larger residue suffices to fill the space close to the ribose ring of the (n – 1) template nucleotide and to restrict the movement of the DNA in the vicinity of the nucleotide binding pocket. The Y391M replacement destabilizes the template strand less than the Y391A replacement, resulting in an enzyme with a reduced rate of DNA synthesis that is still sufficient to maintain replication in vivo. The structural changes caused by the Y391M replacement probably affect affinities of incoming dNTPs but do not alter the polymerase catalytic site. Base selection is affected to a much lesser extent by these structural changes, as indicated by the weak fidelity changes in reversion tests.

Tyr567 is maintained in an infrequent rotamer conformation and probably has contacts with both Try391 and Thr587. The Y391F substitution, which would remove its contact with Tyr567, reduced the rate of DNA synthesis only slightly (Table 2). Thus, the putative hydrogen bond between Thr587 and Tyr567 would be sufficient to stabilize the correct Tyr567 conformation and prevent it from blocking the dNTP binding site. The Y391F enzyme is at most a weak mutator in reversion tests, indicating that the Tyr567 side chain can perform its role in base selection even if the putative hydrogen bond to the hydroxyl group of Tyr391 is disrupted. The van der Waals contacts between the Tyr391 aromatic ring and the ribose moiety of the (n– 1) template nucleotide helps to maintain the template strand in a proper alignment within the active site, and these interactions are essential. However, Tyr391 does not appear to assist in maintaining Tyr567 in the unfavorable rotamer conformation as previously proposed.⁶

Role of Tyr619

Replacing the conserved palm subdomain Tyr619 residue with Phe impairs polymerase activity, although not sufficiently to block growth (Table 1). The mutated enzyme has reduced DNA binding affinity (Figure 3), decreased processivity (Figure 4), and decreased exonuclease activity on both ssDNA and dsDNA substrates (Figure 5). In reversion tests in an Exo⁺ background, Y619F displays close to the parental fidelity for base substitutions, slightly elevated −1 mutagenesis, and substantially decreased +1 mutagenesis (Table 2). The same pattern was observed for the Y619A mutant (data not shown). In the forward-mutation test, the same polymerase increased the overall mutation rate about five-fold (Table 5). In both reversion and forward-mutation tests, the Pol^Y619F Exo⁻ mutant was an approximately five-fold antimutator (Tables 3 and 5).

The rI mutational spectra (Table 6 and Figure 2) show that single-base additions and deletions are infrequent with Pol⁺ Exo⁺ and Pol^Y619F Exo⁺, whereas small additions increase substantially in both the Pol⁺ Exo⁻ and Pol^Y619F Exo⁻ spectra, indicating that proofreading repairs +1 mutations more efficiently than −1 mutations in both enzymes. All the ±1 mutations arise in homopolymeric runs of 4–5 A:T base pairs. The antimutator effect of Pol^Y619F Exo⁺ on +1 mutations observed in the reversion test was absent from the forward-mutation spectrum, indicating that the reversion effect was special for the 131 site. The Pol⁺ Exo⁻ polymerase generates predominantly −1 mutations in vitro and most of these arose in runs of 2–5 repetitive bases (Bebenek et al. 2002).¹⁹ These differences between the in vivo and in vitro data may again be explained by different sequence contexts in the target sequences. For instance, for wild-type T4, the frequency of +1 mutations within runs of the same length (five A:T base pairs) differs at different sites in the rII gene.^20,21

HIV-1 RT and E. coli Pol I Klenow-fragment mutants with decreased processivity due to disrupted polymerase contacts with the DNA backbone upstream of the polymerase active sites displayed increased mutations rates, especially for ±1 mutations in homopolymeric runs.^8,22 For HIV-1 RT mutants, sites of termination of processive synthesis were also sites of elevated frameshift mutagenesis.⁸ A Klenow fragment with a 24-residue deletion at the tip of the thumb (the region that interacts with the template-primer) had decreased DNA binding affinity and reduced processivity and was a strong mutator for indels, especially +1 errors in homopolymeric runs.²³ These results clearly show that altering polymerase-DNA interactions can affect frameshift fidelity and are consistent with the hypothesis that dissociation and reassociation of the enzyme-DNA complex may promote slippage of the template and primer strands and formation and/or utilization of misaligned frameshift intermediate.^24,25

A misaligned +1 mutational intermediate has an extra unpaired nucleotide in the primer strand.^26,27 To generate a +1 mutation, this nucleotide must be stabilized by a correct base pair at the primer terminus. The strong RB69 gp43 Exo activity removes most base-substitution and frameshift errors, especially +1 additions. Pol and Exo activities are coordinated and the primer strand can move between these two sites, with three base pairs melted to access the Exo site.²⁸ The antimutator effect in the reversion assay for +1 mutations may reflect a reduced ability of the mutant polymerase to extend from misaligned substrates. Enzyme with lower DNA-binding affinity will dissociate more frequently from such a substrate, providing the exonuclease increased opportunity to digest the distorted primer and allow the polymerase to rebind and continue synthesis.

We provide some of the structural explanations that may help to understand the unexpected behavior of the RB69Y619F mutant. Tyr619 is located at the protein-DNA interaction surface of the DNA binding groove formed by the thumb and palm subdomains. Tyr619 (palm), Tyr626 (palm), Trp702 (thumb) and Tyr708 (thumb) form a cluster of aromatic residues at the interface of the palm and thumb subdomains. Both Val396 and the aliphatic portion of the Lys705 side chain also contribute to this hydrophobic cluster. Tyr708 and Lys705 belong to the highly conserved KKRY motif. The hydroxyl group of Tyr619 participates in the hydrogen-bond network at the protein-template DNA interface, forming a water-mediated hydrogen bond with the phosphodiester at the primer 3′ terminus, which is also directly hydrogen bonded to Tyr708. The same water molecule also participates in the hydrogen bond with conserved residue Asp621. The Asp621 side chain forms a hydrogen bond with Lys706, the second Lys of the KKRY motif. It has been suggested that the role of the hydrogen bonds formed by residues belonging to KKRY is to draw the primer and template together.¹⁵ Therefore, the Y619F substitution, which perturbs this hydrogen bond network, may change the relative positions of primer and template strands and affect alignment of the primer 3′ terminus at the active site, slowing phosphodiester bond formation and the rate of DNA synthesis both in vivo and in vitro but having little effect on overall fidelity. The Y619F substitution also reduces the proofreading exonuclease activity of RB69. This picture invites detailed kinetic studies of both the parental and the Y619F polymerases.

Materials and Methods

Strains

Escherichia coli BB cells carrying a plasmid expressing a desired allele of RB69 gene 43 were used to grow stocks of T4 43am (in which codons 202 and 386 were converted to TAG) with or without an rII mutation. QA1 cells were used to assay T4 43am rII ⁺ revertants. B40 suII ⁺ cells were used to assay T4 am43 stocks with or without rII mutations. BB cells carrying a plasmid expressing wild-type RB69 gene 43 were used to screen for mutant plaques displaying the phenotype of rI mutants. NR9099 cells were used for growing M13mp2 phage, MC1061 cells were used for electroporation of M13 gap filling reactions, and CSH50 cells were used as the detector strain as described previously.²⁹

Plasmid pCW19R carries a wild-type RB69 gene 43 under the control of the T7 Φ10 promoter of cloning vector pSP72 (Promega). The Exo⁻ mutants carry D222A and D327A replacements. Mutant derivatives of pCW19R were constructed by site-directed mutagenesis and confirmed by sequencing. Enzyme from plasmid pEZ2.1 carrying Y619F in an Exo⁺ background is designated Pol^Y619F Exo⁺; from pAB-1, Pol^Y619F Exo⁻; from pSNG 54.2, Pol^Y619A Exo⁺; from pSNG17-1, Pol ^Y391F Exo⁺; from pSNG18-3, Pol^Y391F Exo⁻; from pCAL51.3, Pol ^Y391M Exo⁺; from pLY391M, Pol^Y391M Exo⁻; from pSNG 23-2, Pol^Y391A Exo⁺; from pSNG, Pol^Y391A Exo⁻; and from pCW.50R, Pol⁺ Exo⁻. All plasmids except pAB-1 were generous gifts from Jim Karam (Tulane University).

Genetical methods

Reversion assays at four different rII sites were as described.^1,2 Stocks of T4 am43 rII mutants were grown in E. coli BB cells expressing a desired allele of RB69 gene 43. rII131 reverts mainly by +1 (A5→A6), rIIUV232 by −1 (A3→A2), rIIUV256 by G:C→A:T transitions, and rIIUV375 by both transitions and transversions at three adjacent A:T sites.

Forward-mutation assays were as described.² rI mutants of independent origin were collected, mutant plaques were resuspended in 40 μl of water, and the rI gene was amplified by PCR and sequenced.

Mutation rates μ in vivo were calculated as μ =f/ln(Nμ) where f is the mutant (or revertant) frequency and N is the final population size.

In vitro gap-filling assays on M13mp2 lacZα substrates were as described.²⁹ The incubation mixture contained 25 mM Tris-acetate, pH7.5, 10 mM (CH₃COO)₂Mg, 150 mM CH₃COOK, 2 mM dithiothreitol, 150 ng M13mp2 lacZα gapped substrate, 1 mM of each dNTP, and 8–23.8 pmol of Pol^Y391A Exo⁺ or 16.3 pmol of Pol^Y391A Exo⁻. Reactions were incubated at 37 °C for 30 min, stopped by addition of EDTA, and analyzed by agarose gel electrophoresis. Products of complete gap-filling reactions were introduced into M1061 cells and plated on CSH50 cells to score M13mp2 plaques as either wild-type (dark blue) or mutant (lighter blue to colorless).

Biochemical methods

SDS PAGE gel-purified oligonucleotides for exonuclease and DNA gel-retardation assays were from SIGMA Aldrich. All primers were 5′ end-labeled with γ-[³²P]ATP (3000 Ci/mmol, Amersham Biosciences) using T4 DNA polynucleotide kinase. Annealing of primer to template was in 10 mM Tris-HCl, pH7.5 with a primer:template molar ratio of 1:1.3 for exonuclease and gel-retardation assays, and 1.2:1 for M13mp2 lacZα assays. The primers were annealed at 70 °C for 5 min and allowed to cool slowly to 25 °C.

Measurements of T4 phage DNA synthesis in vivo were as described² except that cells were grown in LB medium to achieve vigorous growth. [³H] thymidine was at 5 μCi/ml and specific activity 2 mCi/mmol.

cDNAs encoding Pol⁺ and Pol^Y619F polymerases were cloned into the Promega vector SP72. Expression induction with IPTG was in E. coli BL21 Codon Plus RIL cells (Stratagene). After induction, cells were centrifuged, resuspended in buffer B_0.05 (25 mM MOPS, pH 6.8, 10% glycerol, 50 mM NaCl, 5 mM MgCl₂, 0.5 mM benzoamidine, 0.1 mM PMSF, 1 mM EDTA), sonicated, and centrifuged for 20 min at 18000 rpm. Lysate proteins were batch-bound with ~12 ml P11 (equilibrated with the same buffer) for 1 h at 4 °C with constant mixing. The contents of the tube were transferred to an Amersham C16/30 column, which was washed with 30 ml of B_0.05 buffer. Proteins were eluted with a linear gradient of 150 ml 0.05–1 M NaCl. Amounts of proteins were estimated by the Bradford method. Protein in peak fractions was verified by SDS-PAGE electrophoresis. Fractions were pooled and proteins were precipitated overnight with ammonium sulfate at a final concentration of 2.4 M. Precipitated proteins were pelleted at 7000 rpm for 20 min. The precipitate was dissolved in 2.5 ml of buffer A_0.04 (20 mM Tris-HCl, pH 7.5, 5% glycerol, 5 mM MgCl₂, 40 mM NaCl, 1 mM DTT, 0.5 mM benzoamidine, 0.1 mM PMSF, 0.1 mM EDTA). The fractions were desalted through C10 (Amersham) columns equilibrated with A_0.04 buffer. Desalted fractions were loaded onto FPLC MonoQ columns equilibrated with the same buffer. A 30-ml gradient of 0.025–0.2 NaCl was applied, and 2-ml fractions were collected. As judged by ultraviolet transmission and SDS-PAGE electrophoresis, gp43 eluted at about 200 mM. The fractions were pooled and dialyzed overnight against 250 ml of A₅₀ (20 mM Tris-HCl, pH 7.5, 100 mM KCl, 50% glycerol, 10 mM Mg-acetate, 2 mM DTT, 0.1 mM PMSF, 0.1 mM EDTA). Protein concentrations were estimated in a NanoDrop spectrophotometer at OD₂₈₀. Polymerases were stored in small portions at −70 °C.

Exonuclease assays were in the same reaction buffer. The 60-μl reaction mixture contained 25 mM Tris-acetate, pH7.5, 10 mM (CH₃COO)₂Mg, 150 mM CH₃COOK, and 2 mM DTT. The ssDNA assay contained 100 nM of ³²P-labeled 20-mer (5′-TTTGTAGTATTATCAAGTCG-3′) as substrate and 0.5 nM (enzyme:DNA ratio = 1:200) of gp43. The dsDNA assay contained 50 nM of ³²P-labeled 20-mer (5′-TTTGTAGTATTATCAAGTCG-3′) annealed to unlabeled 35-mer (5′-TGCCTTCGTAATCTTACAATTGATAATACATCAAA-3′) and 5 nM of gp43 (enzyme:DNA ratio = 1:10). Reactions were incubated at 37 °C for 10, 20, 40, 80, 180 and 300 sec and 10-μl samples were quenched by adding 5 μl of stop-dye solution. After heating at 95 °C for 5 min, samples were resolved by electrophoreses in 7M urea/15% polyacrylamide gels, analyzed by autoradiography, and quantified by densitometry. For exonuclease assays, nMoles of remaining primer were plotted against time. The binding of gp43 to primer-template substrates was assayed using a 5′-labeled 20-mer (5′-TTTGATGTATTATCAATTGT-3′) annealed to a 35-mer (5′-TGCCTTCGTAATCTTACAATTGATAATACATCAAA-3′). The reaction conditions were as described in³¹ with small modifications. The 20-μl incubation mixture contained 10 mM HEPES, pH 8.0, 0.5 mM dithiothreitol, 1 μg BSA (100 μg/ml), 50 nM of DNA substrate, and 2.5, 5, 10, 25, 50, 100 or 200 nM of gp43. After incubation for 5 min at room temperature, samples were supplemented with 5μl of gel-loading buffer (5 mM HEPES-NaOH, 5% glycerol) and subjected to electrophoresis on a pre-cooled 6% polyacrylamide gel (monomer:bis = 29.5:0.5) containing 20 mM HEPES-NaOH and 0.1mM EDTA. Electrophoresis was run at 4 °C in a running buffer (20 mM HEPES-NaOH, 0.1 mM EDTA) at 110 V. After 2.5 h, gels were dried, exposed on a PhosphoImager screen, analyzed on a Storm 860 PhosphoImager (Molecular Dynamics), and quantified by densitometry.

Polymerase processivities were examined using M13mp2 lacZα ssDNA substrate primed with the 20-mer 5′-TGCAAGGCGATTAAGTTGGG-3′ complementary to nt 85–105 of the lacZα sequence. The 20-mer was phosphorylated at the 5′-end as described above. The 30-μl reaction mixture contained 25 mM Tris-acetate, pH7.5, 10 mM (CH₃COO)₂Mg, 150 mM CH₃COOK, 2 mM dithiothreitol, 150 fmol primer-template substrate, and 4.3 fmol of Pol⁺ Exo⁻ gp43 (primer-template:enzyme ratio = 35:1) or 29.5 fmol of Pol^Y619F Exo⁻ gp43 (primer-template:enzyme ratio = 5:1). Reactions were incubated at 37 °C and 10-μl aliquots were removed after 5, 10 and 15 min and mixed with stop-dye solution. Under these reaction conditions, determined experimentally for each polymerase, the amount of reinitiation on previously extended template-primer molecules relative to the total pool of product DNA was negligible (2%–5%)³². Samples were heated at 95 °C and 5 μl were run on 7 M urea/12% polyacrylamide gels. The DNA products were quantified by phosphorimaging. DNA markers were from dideoxy sequencing on the same substrate.

Abbreviations used

Exo

proofreading exonuclease domain

Pol

polymerase domain

indel

addition or deletion of any number of bases