similar sequence in many eukaryotic and viral. DNA polymerases and in bacteriophage T4 DNA polymerase. We designed genetic techniques to isolate mutant.
THEJOURNAL OF BIOLCGICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 269,No. 8, Issue of February 25,pp. 5635-5643, 1994 Printed in U.S.A.
Motif A of BacteriophageT4 DNA Polymerase: Rolein Primer Extension and DNA Replication Fidelity ISOLATION OF NEW ANTIMUTATOR AND MUTATOR DNA POLYMERASES* (Received for publication, June 30, 1993, and in revised form, October 2, 1993)
Linda J. Reha-KrantzS and RandyL. Nonay From the Department of Genetics, University of Alberta, Edmonton, Alberta T6G 2E9, Canada erases, HIV-1 reverse transcriptase and the Klenow fragment Polymerases in general share only a few regions of amino acid similarity. One of the most conserved re- of E. coli DNA pol I (Kohlstaedt etal., 1992). The structures of gions,called motifA, has the sequence DXXSLYPSII or a both polymerases share common features, particularly in the A C. For both similarsequenceinmanyeukaryoticandviral DNA protein regions that include conserved motifs and residues inmotifs A polymerases and in bacteriophageT4 DNA polymerase. polymerases, highly conserved carboxylate We designed genetic techniques to isolate mutant T4 and C are located near each other in the three dimensional DNA polymerases with amino acid substitutions in thisstructure, although these residues are separatedby 76 amino higldy conserved motif. The mutant DNA polymerases acids for HIV-1 reverse transcriptase and by 185 amino acids of the motifs A differed from wild type T4 DNA polymerase in several for E. coli DNA pol I. In addition, the structures ways. For one mutant DNA polymerase, the pyrophos- and C protein regions of the two polymerases are nearly identical (Kohlstaedt et al.,1992). Despite these elegant structural phate analog, phosphonoacetic acid, was a potent inhibitor of DNA replication, and this mutant DNA polym- studies, theroles of motifs A and C in polymerase function are still unclear, because it has not yet been possible to visualize erase replicated DNA with reduced fidelity. Another mutant DNA polymerase replicatedDNA with increased ac- duplex DNA bound to this region of a polymerase. From the curacy, but this mutant DNA polymerase was less pro- most recent structuralinformation (Beeseet al., 1993)and from cessive in primer extension reactions, and DNA mutational analyses (Polesky et al., 1992),however, these motifs are predicted to be in the active site. replicationrequiredhighconcentrations ofdeoxyWe have designed genetic selection strategies to isolatebacnucleoside triphosphates. We provide evidence that indicates that all of these changes to DNA polymerase teriophage T4 DNA polymerase mutants with amino acid DNA changes in motif A to probe the function of this highly confunction are due to differences in how the mutant polymerasespartitionbetween states active for DNA served polymerase motif. The advantage in using a genetic replication or exonucleolytic proofreading. These stud- approach is that the mutant DNA polymerases selected still ies also provide further support for the hypothesis that retain the ability to replicate DNA but withsignificant changes the accuracy of DNA replication observed for DNA poly- compared with wild type DNA polymerase function. The new merases dependson the interplay between polymerase phenotypes produced by the mutantDNA polymerases provide N., Poland, the means to probe DNA polymerase function in vivo and in and 3‘ -+ 6’ exonuclease activities (Muzyczka, B L., and Bessman,M. J. (1972)J. BioZ. Chern. 247,711G vitro. 7122). We discovered early in ourgenetic studies thatconservative amino acid substitutions within the motif A sequenceproduced significant changes t o DNA replication fidelity. While some muComparative studies of protein sequences of DNA polymertant DNA polymerases replicated DNA with increasedaccuracy ases, DNA-dependent RNA polymerases, RNA-dependent DNA (antimutator phenotype), anothermutant DNA polymerase polymerases, and RNA-dependent RNA polymerases revealed replicated DNA with decreased accuracy (mutator phenotype). two protein motifs, calledA andC, which appear tobe generally In one case, different conservative amino acid substitutions at conserved (Delarue etal., 1990). Even though proteinsequence the same position in the motif produced either mutator or similaritiesbetweenmany polymerases are limited to just antimutator phenotypes. High fidelity DNA replication is these motifs, it was proposed that structural featuresmay nevachieved by accurate nucleotide incorporation followed by a ertheless be conserved and that allpolymerases may resemble proofreading step (notereviews by Kunkel(1992) andby GoodDNA polymerase I (pol I)’ of Escherichia coli (Delarue et al., man etal. (1993)).The interplaybetween polymerase and 3’ + 1990; Joyce, 1991). Recently, experimental supportfor this pro5’ exonuclease activities is critical in determining DNA repliposal was provided by comparing the structuresof two polymcation fidelity (Muzyczka et al., 1972). For E. coli DNA pol I, * This work was supported by grants from the Natural Sciences and active sites for nucleotide incorporation and 3’ + 5‘ exonuclelocated in separate protein domains (Olliset al., Engineering Research Council of Canada, the National CancerInstitute ase activity are of Canada, with funds from the Canadian Cancer Society, and the Al- 1985).It islikely that otherproofreading DNA polymerases are berta Heritage Foundation for Medical Research (to L. J. R.-K.). The also multidomain structures with distinct polymerase and 3’ costs of publication of this article were defrayed in part by the payment 5’ exonuclease active sites (reviewed by Blanco et al. (1991)). of page charges. This article must therefore be hereby marked “aduerThus, DNA polymerases must have some mechanism that altisement” in accordance with 18 U.S.C. Section 1734 solely to indicate lows for the repositioning of the primer terminus between the this fact. $ Scholar of the Alberta Heritage Foundationfor Medical Research. polymerase and 3’ + 5’ exonuclease active sites. We propose To whom correspondence should be addressed. Tel.: 403-492-5383; Fax: that motifAplays a role in determiningif the primer terminus 403-492-1903. The abbreviations used are the following: pol I, E. coli DNA pol I; is a substrate for extension or for 3’ + 5’ exonuclease activity PAA, phosphonoacetic acid; dNTP, deoxynucleoside triphosphate; P T , and, thus, affects DNA replication fidelity. primer template; Cf, final concentration. Most DNA polymerases cannot efficiently extend a mispaired
-
5635
5636
Function Motif of
A of T4 DNA Polymerase
primer terminus, the exception being reverse transcriptases (Mendelman et al. (1990), Yu and Goodman (1992), and note review byKunkel (1992)). Hence, extension ofa mispaired primer terminus will be slow relative to switching from the polymerase tothe 3' + 5' exonuclease active site so that proofreading will be the most likely reaction. If nucleotide incorporation is fast relative to the switch to the 3' + 5' exonuclease active site, as predicted to occur with a correctly base paired primer terminus, then nucleotide incorporation rather than primer degradation will be the favored reaction. The kinetic parameters for T4 DNA polymerase incorporation and proofreading activities have been determined, and these values support the mechanism discussed above (Capsonet al., 1992; Goodman et al., 1993). In support of our proposal, we found that amino acid substitutions within the motif A sequence of T4 DNA polymerase either increased or decreased "switching"between the polymerase and 3' + 5' exonuclease active sites. One mutant DNA polymerase with increased DNA replication fidelity was more frequently in a state active for hydrolysis. Another mutant DNA polymerase with decreased DNA replication fidelity was less frequently in a state active for hydrolysis even when dNTP pools were low.For wild type T4 DNA polymerase, the balance between states active for polymerase or proofreading is intermediate between the two mutants and likely reflects a compromise in the need to maximize DNA replication fidelity while assuring efficient DNA replication without excessive waste of dNTPs. We conclude that amino acid residues in motif A play a critical role in regulating the balance between polymerase and proofreading activities.
al., 1986). DNA polymerase mRNA was sequenced using a modified dideoxynucleotide procedure with avian myeloblastosis virus reverse transcriptase and 14 synthetic oligonucleotide primers complimentary to T4 DNA polymerase mRNA (Reha-Krantz, 1988). Sequencing products were resolved using standard gel electrophoresis methods. DNA Synthesis in V i v e T 4 DNA synthesis rates at 30 "C were determined in vivo for wild type and mutant strainsin E. coli optA+and optAl hosts. Bacteria were grown to 2 x lo8 celldml in M9 minimal medium supplemented with 0.4% glucose, 0.5% casamino acids, and 0.01% MgSO,. T4 phage were added to give a multiplicity of infection equal to 5. At 4-min postinfection, thymidine and 2' deoxyadenosine were added t o final concentrations of 2 and 25 pg/ml, respectively. T4-infected cells were then distributed in 0.5-ml samples to separate tubes, and incubation was continued with aeration by shaking. At 10min intervals, 5 pl of 13H]thymidine(288 pCi/ml) was added. Thirty seconds later, further incorporationwas stopped by addition of 0.2 mlof ice-cold 15%trichloroacetic acid. Acid-insolublematerial was collected on Whatman GF/A filters and counted by standard liquid scintillation procedures. T4 DNA Polymerase Purification-Wild type and mutant DNA polymerases were purified frominducedcells containing the T4 DNA polymerase expression vector (Lin et al., 1987).Modified expression vectors that produced the L412M and I417V-DNA polymerases were provided by T.-C. Lin, I. Douglas, and W. Konigsberg. Detailed T4DNA polymerase purification methods have been described (Reha-Krantz et al., 1993). Briefly, homogenous protein, as judged by SDS-polyacrylamide gel electrophoresis, was recovered in five steps: 1) cell lysis; 2) partial removal of nucleic acids by streptomycin sulfate precipitation; 3) (NH,),SO, precipitation; 4) fractionation over Q-Sepharose(Pharmacia LKB BiotechnologyInc.); 5) fractionation over P-11 (Whatman). T4 DNA polymerase concentration was measured spectrophotometrically using the experimentally determined molecular extinction coefficient at 280 nm (EM,,)of 1.49 x lo5 M - ~cm" (Reha-Krantz et al., 1991). DNA Polymerase Assays-DNA polymerase activity was measured under two different reaction conditions. One reaction condition was a 60-pl volume reaction with 67 m~ Tris-HC1 (pH 8.8), 16.7 m~ MATERIALSANDMETHODS (NH.&S04, 0.5 m~ dithiothreitol, 6.7 m~ MgCI,, 167 pgml bovine seBacterial and T4 Bacteriophage Strains andMicrobiological Proce- rum albumin, 83 p~ dNTPs (one labeledL3H1dNTP at -100 cpm/pmol), dures-The E. coli strain CR63 was used for preparation of T4 phage and either 833 p~ "activated" DNA (expressed in nucleotide equivcultures as describedpreviously (Reha-Krantz and Lambert, 1985). alents, preparation describedbelow) or 0.83 AZm unitdml polyCR63 was also the host used to detect phosphonoacetic acid (PAA) (dC).oligo(dG)lz-18(Pharmacia, annealed at a dCMP:dGMP ratio of 1) inhibition of T4 phagegrowth. PAA sensitivity was determined by com- or 0.83 unitdml poly[d(A-T)] (Pharmacia). In reactions with wild paring plating efficiencies of phage on standard Hershey broth plates type T4 DNA polymerase, 40 ng of protein was added to start the and on plates supplemented with 1mg/ml PAA (Sigma)(Reha-Krantzet reaction, which produced a linear nucleotide incorporation rate for a t al., 1993). Wild type plating efficiency was reduced about 40% under least 30 min at 30 "C. Reactions were stoppedby spotting 20 pl of the these conditions. reaction mixture onto GF/A filters (Whatman), which were prespotted Some mutant T4 phage strains are restricted on the E. coli optAl with 20 pl of 0.1 M EDTA. The filters were floatedin an ice-cold solution host (Gauss et al., 1983). OptAl and optA' strains, NapIV 2396 and of 0.1 M NaPPi and 7.5%trichloroacetic acid. Afterall the samples were 2395 strains, respectively, were providedby P. Gauss. Phage growth in collected, the filters were washed under vacuum with cold water, dried, optA1 cells compared with optA+ cellswas determined by measuring the and counted in scintillation fluor. number of phage produced per bacterium in a single growth cycle at Activated DNA was prepared by partially digesting salmon sperm 40 "C (burst size). Wild type T4 phage produced about the same number DNA with DNAse I by the method of Oleson and Koerner (1964) as of progeny in both cell types, while some T4 DNA polymerase mutant modified by Reha-Krantz and Nonay (1993).One unit of DNA polymerstrains produced essentially no progeny in the o p a l host. ase catalyzes the incorporation of 10 nmolof labeled nucleotideinto an Spontaneous mutation frequencies were determined by measuring acid-insoluble product in 30 min at 30 "C. the number of revertants at anrII site, rW199, asdescribed previously The second type of reaction condition wasa modification of the assay (Reha-Krantz et al., 1986). Strains with DNA polymerase mutations described by Reddy et al. (1992).A 20-pl reaction mixture contained 25 that increase DNA replication fidelity (antimutator phenotype) produce m~ HEPES pH 7.5, 60 m~ NaOAc, 1m~ dithiothreitol, 0.5 m~ EDTA, fewer revertants than observed for wildtype cultures, and strains with 80 p~ dNTPs, 0.2 mg/ml bovine serum albumin, 7.5 I" PT (expressed as mutations that decrease DNA replication fidelity (mutator phenotype) 3' primer ends, the preparation of the PT is describedbelow), and produce more revertants. T4 strains with various DNA polymerase 30-150 I" wild type or mutant T4 DNA polymerase. Excess DNA pomutations were crossed tothe rW199strain t o produce doublymutant lymerase to primer ends was used because of nonspecific binding of DNA polymerase rW199 strains. Five or morehigh titer cultures were DNA polymerase to single-stranded DNA. Except where otherwise grown for each strain, and the number of rW199' revertantdculture noted, DNA polymerase waspreincubated in the above reaction mixture was determined by titering phage on the CR63 bacterial host and the A at 30 "C for 5 min: Reactions werestarted by the addition of Mg" (C, lysogenic form(CR63A) ofthis host that restricts rZI mutant phage. T h e 6 m~ Mg(OAc),) for multiple encounter reactions or by the addition of number of phagdculture that can grow onthe A lysogenic host gives the Mg2+plus heparin (Sigma, Cf at 0.1 mg/ml) for single encounter reacnumber of rW199' revertants. For each group, the culture with the tions. Reactions were stopped by the addition of 2 1.1 of 0.2 M EDTA. median number of rW199+ revertants was chosen as the representaReaction products were separated on DNA sequencing gels (7% acryltive culture for determining the reversion frequency. The values re- amide, 8 M urea), and the32P-labeled reaction products were visualized ported are relative to the wild type reversion frequency of 1.2 rW199' by autoradiography with Kodak X-Omat AR film. revertantd106 phage. PTs were prepared by annealing an oligonucleotide (2l-mer) to a T4 DNA Polymerase Sequence Determinations-Nucleotide se- 6-kilobase single-stranded, circular DNA with a single complimentary quences of wild type and all mutant DNA polymerase genes were de- region and then using an exonuclease-deficientT4 DNA polymerase to termined by sequencing DNA polymerase mRNA produced either from extend the primer. T4 infections or from induced plasmidexpression vectors. Total nucleic acid was purified from T4-infected cells or from DNA polymerase extemplate: 3'-GTCAGCCTTTATTGTAACACTTGCA..(plasmid DNA) 5"CAGTCGGAAATAACATTGTTG (21-mer) pression plasmid cells accordingto published procedures (McPheeterset primer:
5637
Function of Motif A of T4 DNA Polymerase TABLEI Comparison of wild type and mutant T4 DNA polymerasg strains with mutations that encode amino acid substitutions in motif A Phenotypesb Strain
Wild type I417V L412M + I417V L412M L412I S411T + L412M
of motif A"
sequence Amino acid
408DL T S L Y P S I M M I
1417
V V
0.09
TM
OptAl sensitivity
PPP sensltlvity
0.85 0.005 0.90 0.80 0.001 0.005
0.60 0.90 0.002
