Primer-terminus Stabilization at the $29 DNA

THE JOWALOF BIOLCGICAL CHEMISTRY 0 1994 hy The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 47, Issue of November 25, pp. 30030-30038, 1994 Printed in U.S.A.

Primer-terminus Stabilization at the$29 DNA Polymerase Active Site MUTATIONAL ANALYSIS OF CONSERVED MOTIF TX,GR* (Received forpublication, June 7, 1994, and in revised form, September 9, 1994)

Juan MBndezS, Luis Blanco, Jose M. Lazaro, and Margarita Salasg From the Centro de Biologia Molecular “Seuero Ochoa” (C.S.I.C.-U.A.M.), Uniuersidad Aut6noma, Cantoblanco, 28049 Madrid, Spain

Thefunctionalsignificanceoftheconservedmotif several segmentsof amino acid similarity have been identified TX,GR, included in one of thesix main regions of amino in the C-terminalportion of eukaryotic typeDNA polymerases acidsequencesimilarityidentifiedintheC-terminal (Bernad et al., 1987; Larder et al., 1987; Wong et al., 1988; portion of bothEscherichia coli DNA polymerase I-like Blanco et al., 1991) and prokaryotic, pol I-like DNA polymandeukaryotic-type DNA polymerases (Blanco, L., erases (Ollis et al., 1985; L6pez et al., 1989; Lawyer et al., 1989; Bernad, A, Blasco, M A , andSalas, M. (1991) Gene Leavitt and Ito, 1989). Alternative alignments of these con(Amst.) 100,2748) has been studied by site-directed muserved C-terminal segments between thetwo families of DNA tagenesis in the 429DNA polymerase.A revised multiple polymerases have been proposed (Matsumoto et al., 1989; Dealignment of this region, including 61 DNA polymerases larue et al., 1990; Blanco et al., 1991; Grabherr et al., 1992). belonging to these two superfamilies, is presented. In The role of particular residues belonging to the most conaddition, based on amino acid sequence comparisons and by extrapolation to the crystal structure T7 of RNA served C-terminalmotifs has been studiedby site-directed mupolymerase, a similar motif (DX,GR)is predicted to be tagenesis in 429 DNA polymerase (reviewed by Blanco and structurally and functionally equivalent RNA in polym- Salas (1994)) and other DNA-dependent DNA polymerases. erases, the other class of DNA-dependent polymerases. Thus, motifYGDTDS is probably involved in metal binding and et al., 1990), herpes (Dorsky and The severe defect in polymerization displayed by twocatalysis in 429 (Bernad of the 429 DNA polymerase mutants used inthis study Crumpacker, 1990; Marcy et al., 1990), adenovirus(Joung (T434N and R438I)is interpreted as the consequence of et al., 1991), and human a (Copeland and Wang, 1993) DNA polymerases. Motif DX,SLYP has been shown to be important a decreased capacity to stabilize the binding of primer(Blasco et al., 1993a), template DNA structuresin a polymerization-competent for dNTP binding and/or catalysis in 429 T4 (Reha-Krantzet al.,19931, and humana (Dong et al., 1993a, conformation.Thesemutantswerealsoseverelyaffected in the formation of terminal protein (TP)-dAMP 1993b) DNA polymerases. Residues of the motif KX,NSXYG initiation complex, a reaction in which 429DNA polym- have been shown toinvolved in primer-template binding and be erase is able to use the TP as primer. dNTP positioning in 429 DNApolymerase(Blasco et al., 1992a, 1993b). Motif KXY has been proposed t o contribute to primer binding in 429 DNA polymerase (Blasco et al., 1992b). This Although DNA polymerases constitute a wide group of en- proposal has been recently supported by analysis of highly zymes of prokaryotic and eukaryotic (viral or cellular)origin, purified mutant proteins in both Lys and Tyr residues.’ several genetic and biochemical evidences led to the proposal Motif TX,GR is present in both groupsof prokaryotic type that all DNA polymerases may have a similar structure. This and eukaryotic type DNA polymerases (Blanco et al., 19911, claim has been supported in part by the study of the three- except in the subgroupof TP-primed DNA polymerases, which dimensional structures of the Klenow fragment of pol I (KF)’ contain a less strict motif, TX,AR. Whereas the Arg residue andhuman immunodeficiency virus type I reversetranbelonging t o the TX’GR motif in KF has been shown t o be scriptase (reviewed by Joyce and Steitz (1994)). The proposed important for DNA binding andcritical for the catalysisof the underlying similarity between all DNA polymerases has been reaction (Polesky et al., 1990, 19921, no site-directed mutagenalsosupported by sequence analysisand site-directed mu- esis studies in this motif of any eukaryotic type DNA polymtagenesis studies. erase have beendescribed so far. In addition to the generally conserved Klenow-like 3‘-5‘In this paper, we report the characterizationof purified 429 exonuclease active site, located at the N-terminal portion of DNA polymerase mutant derivatives T434N, A437G, R438K, DNA polymerases (Bernad et al., 1989; Blanco et al., 1992a), and R438I in motif T X F . The results obtained indicate that mutants in residues Thr434and Arg438(i) havea decreased abilGrant 5R01 *This investigation was supported by Research ity to binda primer-template DNA structure; (ii)are partially GM27242-15 fromthe National Institutes of Health, GrantPB90-0091 affected in TPbinding; and (iii) could be implicated, directly or from Direcci6n General de Investigaci6n Cientifica y TBcnica, Grants of the hypothesis that a BIOT CT 91-0268 and CHRX-CT93-0248 fromthe European Economic indirectly, in catalysis. In support could be conserved in all nucleic acid polymCommunity, and an institutional grant from Fundaci6n Ram6n Areces. similar active site The costs of publication of this article were defrayed in part by the erases, we further report the identification of a similar motif payment of page charges. This article must therefore be hereby marked (DX’GR) in monomeric and multimeric DNA-dependent RNA “aduertisement”inaccordancewith 18 U.S.C.Section 1734 solely to polymerases. indicate this fact. $ Recipient of a predoctoral fellowship from Ministerio de Educaci6n MATERIALS AND METHODS y Ciencia. (400 CiJmmol)and[y-32P1ATP (5000 CiJ Nucleotide~-[a-~~P]dATP 0 To whom correspondence should be addressed. International Plc. Unlabeled nuThe abbreviations used are: KF, Klenow fragment of pol I; dNTP, mmol) were obtained from Amersham deoxynucleoside triphosphate; PAGE, polyacrylamide gel electrophoreM. A. Blasco, J. MBndez, J. M. Lbzaro, L. Blanco, and M. Salas, sis; pol I and pol 11, DNA polymerase I and 11, respectively, from E. coli; TP, 429terminal protein;TP-DNA, 429terminal protein-DNAcomplex. submitted for publication.

30030

429 DNA Polymerase Mutations Affecting Primer Stabilization cleotides were from Pharmacia Biotech Inc. Proteins-Restriction endonucleases and T4 polynucleotide kinase were purchased from New England Biolabs. Wild-type429 DNA polymerase was purified as described (Lbzaro et al., 1994). $29 DNA polymerase site-directed mutants were purified essentially in the same form, from Escherichia coli BL2l(DE3)pLysScells (Studier andMoffatt, 1986) harboring the corresponding recombinant plasmids. 429 TP was purified as described (Zaballos et al., 1989). DNA Templatesand Substrates-TP-DNA was obtained as described (Peiialva and Salas, 1982). 429 DNA was obtained by proteinase K treatment of phage particles in the presence of SDS (Inciarte et al., 1976), followedby phenol extraction and ethanol precipitation. $29 DNA was digested with EcoRI to generate DNA fragments with 3'recessive ends, suitable as templates for DNA polymerization assays (filling in). M13mp8 single-stranded DNA was hybridized to the universal primer, and the resulting molecule was used as a primer/ template in the processive DNA replication assay. Oligonucleotidesspl (5' GATCACAGTGAGTAC) and splc+6 (5' TCTA'M'GTACTCACTGTGATC) were prepared using a 380A synthesizer (Applied Biosystems) and purified by electrophoresis on 8 M urea, 20% PAGE.Oligonucleotide spl was 5'4abeled with [y-3zP]ATPand T4 polynucleotide kinase and further purified by PAGE. The specific activity obtained was 5.5 x lo5 cpm/pmol. Labeledspl was used as substrate for the 3'-5'-exonuclease activity. Labeledspl and unlabeled splc+6 were hybridizedin the presence of 0.2 M NaCl and 60 mM Tris-HC1, pH 7.5. Site-directed Mutagenesis and Expression of 429 DNA Polymerase Mutants-Thewild-type 429 DNA polymerase gene cloned into M13mp18 was used for site-directed mutagenesis carried out as described (Nakamaye and Eckstein, 1986) using the oligonucleotide-directed in vitro mutagenesis kit from Amersham International Plc. The fragments carrying the different mutations were subcloned in plasmid pT74w2 (LBzaro et al., 1994), which expresses 429 DNA polymerase under the control of the T7RNA polymerase-specific 410 promoter (Tabor and Richardson, 1985). The presence of the desired mutations and the absence of any other changes were confirmed by complete sequencing of each 429 DNA polymerase mutant gene. Sequencing was carried out by the chain termination method (Sanger et al., 19771, using Sequenase version 2.0 from U.S. Biochemical Corp. and a set of synthetic oligonucleotidescomplementaryto the 429 DNA polymerase gene as sequencing primers. Expression of the mutant proteins was carried out in the E. coli strain BL2l(DE3)pLysS(Studier, 1991). 3'd'-ExonucleaseAssay-The reaction mixture contained, in 12.5 pl, 50 nm Tris-HC1,pH 7.5, 1 nm dithiothreitol, 4% glycerol, 0.1 mg/ml bovine serum albumin, 10 nm MgCl,, 0.3 ng of a 5"labeled oligonucleotide (spl), and 10 ng of either wild-type or mutant 429 DNApolymerase. After incubation for 1 min at 25 "C, the reactions were quenched with 3 pl of loading buffer and analyzed by 8 M urea 20% PAGE and autoradiography. 3'-5'-exonuclease activity is detected as a decrease in the size of the 5'-labeled DNA substrate. When required, and in order to remove contaminant nucleases, the DNA polymerase fractions were further purified by glycerol gradient sedimentation. Nonprocessive DNA Polymerase Assay(Filling in Reaction)-The incubation mixture contained, in 25 pl, 50 mM Tris-HC1, pH 7.5, 1 mM dithiothreitol, 4% glycerol, 0.1 mg/mlbovine serum albumin, 10 mM MgCl,, 0.1 p [(r-32PldATP(1pCi), 0.2 p each dGTP and d'M'P, 0.25 pg of EcoRI-digested 429 DNA (1.6 pmol of DNA ends) as template, and 10 ng of the correspondingDNA polymerase. Afterincubation for 15min at 30 "C,the reactions were stopped by adding 10 mM EDTA, 0.1% SDS, and the samples were filtered through Sephadex G-50 spin columns in the presence of 0.1% SDS. The excludedvolumewascounted (Cerenkov radiation) and analyzed by agarose gel electrophoresis and autoradiography. The dAMP turnover coupled to the filling in assay (calculated as the ratio between released dAMP and incorporated U P ) was determined by polyethyleneimine-cellulosethin layer chromatography (Polygram Cel300PEIlUV254) and further autoradiography of samples withdrawn immediately after the DNA polymerization reaction. The chromatogram was developed with 0.15 M lithium formate, pH 3.0, conditions in which it is possible to separate 5'-dAMP fromthe DNA and from the unincorporated dNTP that remain at the origin. Polymerasel3'd'-Exonuclease Assay (pol lexo)-The hybrid molecule spl/splc+6 (described above) contains a 6-nucleotide-longB'-protruding end, and therefore can be used both as substrate for the 3"s' exonuclease activity and as primer-template for DNA polymerization. The reaction mixture contained, in 12.5 pl, 50 mM Tris-HC1, pH 7.5, 1nm dithiothreitol, 4% glycerol, 0.1 mg/ml bovine serum albumin, 10 nm MgCl,, 0.3 ng of the hybrid molecule, 10 ng of wild-type or mutant 429 DNA polymerase, andthe indicated concentration of the four

30031

dNTPs. ARer incubation for 5 min at 30 "C, the reactions were stopped by addition of EDTA to 10mM. Samples were analyzed by 8 M urea 20% PAGE and autoradiography. Polymerization orexonuclease are detected as an increase or decrease, respectively, in the size of the 5'labeled primer-strand (spl). of Primed M13 Processive DNA PolymeraseAssay(Replication DNA)-The incubation mixture contained, in 25 pl, 50 mM Tris-HC1, pH 7.5, 1 nm dithiothreitol, 4% glycerol, 0.1 mg/ml bovine serum albumin, 10 mM MgCl,, and 20 p~ each of the four dNTPs in addition to 0.25p~ [ ( U - ~ ~ P ] ~(2.5 A TpCi), P 0.25 pg of primed M13mp8single-stranded DNA (described above), and 10 ng of the indicated DNA polymerase. After incubation for the indicated time at 30 "C,the reactions were stopped by adding EDTA to 10 nm and SDS to 0.1%; the samples were filtered through Sephadex G-50 spin columns in the presence of 0.1% SDS, and the Cerenkov radiation in the excluded volume was counted. DNA Gel Retardation Assay-A hybrid molecule of 5"labeled oligonucleotide spl and oligonucleotidesplc+6 was used to analyze interaction of the wild-type or mutant 429 DNA polymerases with DNA. The incubation mixture contained, in 20 pl, 12 mM Tris-HC1, pH 7.5, 1 mM EDTA, 20 mM ammonium sulfate, 10 mM MgCl,, 0.1 mg/ml bovine serum albumin, 0.072 ng of the hybridmolecule spl/splc+6, and the indicated amount of wild-type or mutant 429 DNA polymerase. Samples were incubated for 5 min at 4 "C and then analyzed by 4% PAGE in the presence of 12 nm Tris-acetate, pH 7.5, and 1mM EDTA. After electrophoresis at 200 V in thesame buffer during 1h at 4 "C, gels were dried and autoradiographed. In the autoradiograms, 429 DNA polymerase-DNA complexesare detected as a shift (retardation) in the position of the labeled DNA. TP-dAMPFormation (Protein-primedInitiation Assay)-The incubation mixture contained, in 25 pl, 50 m Tris-HC1, pH7.5,lO mM MgCl,, 1mM dithiothreitol, 4% glycerol,0.1 mg/ml bovine serum albumin, 0.25 p [CY-~~PI~ (2.5 A TpPa ) , 0.5 pgof TP-DNA, 25ng of purified TP, and 10 ng of the wild-type or mutant 429 DNA polymerase. In the case of the template-independent initiation assay, TP-DNA wasomitted and 1mM MnCl, was used as metal activator instead of MgCl,. After incubation for the indicated time at 30 "C (in conditions shown to be linear with time and enzyme amount), the reactions were stopped by adding EDTA to 10 mM and SDS to0.1%. The samples were filtered in Sephadex G-50 spin columns in the presence of 0.1% SDS. The excluded volume was analyzed by SDS-PAGE and autoradiography. Quantitation was done by densitometric analysis of the autoradiographs. Interference Assay for TP Binding-Reactions were carried out as described forthe protein-primed initiation assay, in the absence of template and using a limiting amount of TP and different proportions of wild-typelmutant DNApolymerases. In the case of mutant D249E, used as a competition control, the amounts of proteins used were as follows: 12.5 ng of TP, 25 ng of wild-type DNA polymerase, and 25 or 100 ngof mutant D249E. For mutant T434N, the amounts of proteins used were as follows: 12.5 ngof TP, 25 ng of wild-type DNApolymerase, and 12.5, 25, 50, or 100 ng of mutant T434N. In both cases, incubation was for 3 h at 30 "C. In the case of mutants R438K and R4381, the amounts of proteins used were as follows: 2.5 ng ofTP, 5 ng of wild-type DNA polymerase, and 5, 10, 15, or 20 ng of mutant proteins. Incubation was for 15 h at 30 "C. In all cases, after incubation, the samples were stopped and analyzed as indicated for the initiation reaction. TP-DNA Replication (Protein-primed Initiation Plus Elongation)The incubation mixture contained, in 25 pl, 50 nm Tris-HC1, pH 7.5, 10 nm MgCl,, 20 nm ammonium sulfate, 1 mM dithiothreitol, 4% glycerol, 0.1 mg/ml bovine serum albumin, 0.5 pg of TP-DNA, 20 p each dCTP, dGTP, and dTTP, and [ C K - ~ ~ P I ~(1ApCi), T P 25 ng of purified TP, and 10 ng of the corresponding DNA polymerase. After incubation for the indicated times a t 30"C, the reactions werestopped and the samples were filtered as described above.The Cerenkov radiation in the excluded volumewas counted and used for quantitation. For qualitative analysis, the labeled DNA was denatured by treatment with 1M NaOH and subjected to electrophoresis in alkaline 0.7% agarose gels. After electrophoresis, the position of unit-length $29DNA was detected by ethidium bromide staining, andthenthe gelswere dried and autoradiographed. RESULTS

The TX,GR Motif of DNA-dependentDNA Po1ymerases"The consensus sequence TX,GR is included in one of the most significant homology regions present in both superfamilies of pol I-like and eukaryotic type DNA polymerases (Blanco et al., 1991). Fig. 1 shows an updated multiple alignment of this region, in which 61 sequences of DNA polymerases belonging to

429 DNA Polymerase Mutations Affecting Primer Stabilization

30032

tttttt

7 -

ttttttt

8 -

EC Pol I /652/ /450/ /413/

/304/ /574/ MiPl

/591/

HSV-1

/825/ /826/

EqW-1

EN- 6 EN-Sai

/680/

/684/ /E251 /730/ /695/ vzv /661/ FPV /661/ Vaccinia / 6 5 0 / ASFV /644/ T4 /571/ CCfV /768/ C.bie. /599/ NF2A /599/ PBCV-1 /599/ / 6 16 / AcMnPV LdnPV /634/ Ec Pol I1 /507/ Vent /504/ PfU /502/ S.solfata /617/ HCMV

MCMV EBV

B Rev3 (Sc) CL (Sc) a (SP) CL (Tb)

a

(Dm)

‘E ?

/E581

/974/

/964/ (Hs)

1 .# E

/1101/ /958/ /944/

Adeno-2 Adeno-7 Adeno-12 DGKLl bGKL2 pCLKl pSKL

s-1

/711/ /777/ /709/ /BOO/

/786/ /845/ /791/ /633/ /966/ /671/

PA12 pMC3-2 /586/ p m maranhar/767/ kalilo /787/ px2-1 /810/ /356/PRDl ;397/ /400/

-

.3

J :ig

* *

FIG.1. The TX,GR motif of DNA-dependent DNA polymerases. The multiple alignment of the amino acid sequences containing motif TX,GR corresponds to one of the most conserved regions within DNAdependent DNA polymerases, originally defined as region 2b (Blanco et al., 1991). Group A : pol I-like DNA polymerases, including bacterial, phage, and mitochondrial enzymes; group B: eukaryotic type DNA polymerases, including bacterial, viral, and cellular enzymes and those that are able to use a protein as primer. Based on KF structural data (Ollis et al., 1985), motif TX,GR would be located in the turn between p-strands 7 and8. DNA polymerase nomenclature and sequence references are compiled in Braithwaite and Ito (1993),with the exception of African swine fever virus DNA polymerase ( A S W , Rodriguez et al. (1993)), DNA polymerase 6 frommouse (6 (Mm); Cullmann et al. (1993)), DNA polymerase E from S. cereuisiae ( E (Sc); Morrison et al. (1989)),DNA polymerases codified by linear plasmids from Marchella conica (pMC3-2;Rohe et al. (1991)) and Podospora anserina (pAL2-I; Hermanns and Osiewacz (1992)),and DNA polymerases fromChlorella viruses, strains N Y 2 A (SwissProt data base, accession number PB30320) and PBCV-I (SwissProt data base, accession number PB30321). Numbers between slashes indicate the amino acid position relative to the N terminusof each DNA polymerase. Numbers in parenthesis indicate the amino acid residues forming a specific insertion in the subgroup of protein-priming DNA polymerases (indicated as SI 2 in Fig. 6B). Starsindicate pol I residue Arp,involved in DNA binding and catalysis (Polesky et al., 1992), the location of HSV-1 DNA polymerase mutations showing altered sensitivity to phosphonoacetic acid,

both main superfamilies (A and B, according to Braithwaite and Ito (199311, have been included. For the multiple alignment, family B (eukaryotictype DNA polymerases) was divided (18 viral sequences, intothree groups: viralandbacterial E. coli pol 11, and 3 DNA polymerases from Archaeobacteria), cellular DNA polymerases (13 sequences), and DNA polymerases that initiate by protein priming (17 sequences). Argresidue is almost invariant in all groups of DNA polymerases (55 sequences out of 61). In KF, thisresidueis shown by site-directed mutagenesis tobe important for DNA binding and critical for the catalysis of the polymerization reaction (Polesky et al., 1990, 1992). Thr is conserved in both groups of pol I-like and eukaryotic type DNApolymerases (46 sequencesout of 61). Gly is conserved in both groups, except in the case of DNA polymerases that use proteins as primers,where this Gly residue is always substituted by Ala, Ser, or Thr. Interestingly, DNA polymerases Chlorella viruses NY-2A and PBCV-1 have the sequence VTATGR, identical t o that inpol I (see Fig. 1).To estimate the significance of the alignment shown in Fig. 1, we bases. Out of used the sequence VTATGR to search protein data 31808 proteins searched, thissequence was only found in four proteins: the three polymerases mentioned above and an ATP synthase from Annabaena sp. According to thethree-dimensional structure of the KF (Ollis et al., 19851, motif TX,GR would be located on the turnbetween p-strands 7 and 8, being part of the polymerization active site. Furthermore, asindicated in the alignment shown in Fig. 1,the residues preceding those forming p-strand 7 in pol I also show traces of similarity with those preceding p-strand 8 in pol I (involving motif TX,GR). This apparent structural duplication can also be observed inother DNA polymerasesequences aligned in Fig. 1. In the case of protein-primed DNA polymerases, a specific insertion preceding the region aligned with pol I P-strand 7 has been previously reported (Blanco et al. (1991); see also Fig. 1). Construction of Site-directed Mutants in the TX,GR Motif-To analyze the effect of mutations in each of the three most conserved amino acid residues of this region in 429DNA polymerase (Thr434, Ala437,and A r & 3 8 ) , single changes were designed takinginto accountsecondary structure predictions (Chou and Fasman, 1978; Garnier et al., 1978) and general suggestions for conservative substitutions (Bordo and Argos, 1991). Mutants T434N, A437G, R438K, and R438I were obtained. The change of Thra4 into Asn was designed to remove the hydroxyl group of the side chain. The change of Ala437into Gly restored theconsensus motif TX,GR, present inmost DNA of polymerases from both superfamilies (seeFig. 1).In the case A r g 4 3 8 , corresponding to the most conserved residue of this motif, one of the mutations was designed to maintain thepositive charge (R438K), whereas the other introduced a hydrophobic side chain (R4381). Site-directed mutagenesis and protein purification werecarried outas described under “Materials and Methods.” 3”5’-ExonucLease Activity of 429 DNA Polymerase Mutants-In agreement with the proposed location of the 3’5’-exonuclease active site of 429 DNA polymerase in the Nterminal portion (residues 1-19013 of the protein (Bernad et ul., 1989; Blanco et al., 1991), the four mutant derivatives in con~~~

L. Blanco, L. Villar, J. M. Lazaro, A. Zabellos, and M. Salas, manuscript in preparation. acyclovir, and aphidicolin (Larder et al., 1987; Knopf, 1987; Gibbs et al., 1988), and the 429 DNA polymerase residues studied in this paper. Relevant amino acid similarity among the different groups is indicated in white letters over a black background. The following conservative amino acids (in one-letter notation) were considered:S and T; A and G; K,R,andH;I,L,M,C,V,Y,andF.

-

429 DNA Polymerase Mutations Affecting Primer Stabilization A wild-type

30033 B T434N

4-21 +21

+I5

FIG.2. DNA polymerasdexonuclease coupled assay. The assays were carried out as described under "Materials and Methods." u s h e "'P-labeled hvbrid molecule spllsplc+6 as"primer/tempke DNA and the indicated concentration of each dNTP. The DNA polymerase used in each case is indicated. Arrows indicate the 15-mer position (nonelongated primer) and 21-mer position (completely elongated primer).

-

J

__

.. .

-

20 40 2 m 4002000 40

4

dNTPs. n M

c

K438K

served residues Thr434,Ala437,and had essentially the wild-type level of 3'-5'-exonuclease activity, measured on single-stranded DNA as substrate (notshown). DNA-primed Polymerization Activity of Mutant $29 DNA Polymerases-The effect of the mutations on the ability of the enzyme to catalyze DNA polymerization onto a DNA primer molecule was first analyzed in a short polymerization assay (filling in of 629 DNA-EcoRI ends) as described under "Materials and Methods." In this assay, the enzyme catalyzes the incorporation of few nucleotides onto a DNA primer, and polymerization properties such as strand displacement and processivity are not required. On the other hand, the analysis of the 3'-5'-exonucleolytic release of dAMP coupled to thisfilling in reaction (turnover) allows us to estimate the stability of incorporated nucleotides. As shown in Table I, mutants T434N and R438I were almostinactivein DNA polymerization, whereas mutants A437G and R438K showed 80 and 70%, respectively, of the activityof the wild-type DNA polymerase. The analysis of the dAMP release during this shortpolymerization assay revealed that proteins T434N and R4381 have about 5and 7-fold higher turnover, respectively, than that of the wildtype enzyme. This phenotype is generally interpreted as the consequence of either aninefficient translocation or a defective stabilization of the primer terminus. Using a 5"labeled oligonucleotide (15-mer) hybridized t o a larger template oligonucleotide (21-mer), it is possible t o analyze the equilibrium between exonucleolysis and polymerization, since the products of the two activities on this primer-

20 40 200 4002nnn 4000

dNTPs. n M

D 4-21

R4381 4-

21

template structure can be simultaneously detected. By addition of increasing amounts of dNTPs, this equilibrium is displaced toward synthesis, exonucleolysis being competed by DNA polymerization. Whereasmutant polymerases R438K (Fig. 2, panel C) and A437G (not shown) had the same dNTPrequire. ment for polymerization as the wild-type DNA polymerase (panel A), mutant proteins T434N (panel B ) and R438I (panel D ) needed a 20-fold greater dNTPconcentration than thewildtype enzyme for full polymerization. The processivity and strand displacement capacity of the mutant derivatives duringDNA polymerization were analyzed in a primed M13 DNA replication assay. As shown in Table I, mutant A437G had anactivity slightly higher than thatof the wild-type enzyme, and mutantR438K was partiallyactive. Size analysis of the reaction products by electrophoresis in alkaline gels revealed that these mutations are not affecting processivity and strand displacement (results not shown). On the contrary, mutant proteinsT434N and R438I were almost inactive (see Table I) in spite of the fact that the dNTPconcentration used was 5-fold higher than that required to complete replication in a pollex0 coupled assay. Gel Retardation of Primer-template DNA Molecules by Wildtype and MutantDNA Polymerases-The binding of the different mutant derivatives of the 429DNA polymerase to a primertemplate DNA (spl/splc+6) was analyzed by gel retardation assays, as described under "Materials and Methods." Mg2' ions strongly stimulate the formation of an enzyme-DNA complex that is competent for DNA polymerization, giving rise to a

$29 DNA Polymerase Mutations Affecting Primer Stabilization

30034

TABLEI Enzymatic acitivities ofthe mutant derivatives $29 of DNA polymerase Numbers indicate percentages of the activity obtained with the wild-type DNA polymerase. A qualitative estimate of either DNA or TPbinding is represented from - (not detected) to +++ (wild type). In thepoVexo coupled assay, the dNTP concentration necessary to complete exonucleolysis and totally replicate the template DNA is indicated. Activity assay

429 DNA polymerase

DNA Wt"

DNA as primer Filling in Turnover poVexo coupledb Processive elongation DNA binding TP as primer TP-DNA replication TP-dAMP formation TP binding a

100 100 0.2 100

EcoRI ends

spl/splc+6' Primed M13 DNA spl/splc+B TP-$29 DNA TP-$29 DNA No template No template

T434N

2 480 4

2.5

+++

+

100 100 100