Characterization of a 3''----5'' exonuclease activity in the phage phi 29-encoded DNA polymerase.

Purified protein p2 of phage phi 29, characterized as a specific DNA polymerase involved in the initiation and elongation of phi 29 DNA replication, contains a 3'----5' exonuclease active on single-stranded DNA, but not on double-stranded DNA. No 5'----3' exonuclease activity was found. The 3'----5' exonuclease activity was shown to be associated with the DNA polymerase since 1) the two activities were heat-inactivated with identical kinetics and 2) both activities, present in purified protein p2, cosedimented in a glycerol gradient.     

Processive replication of single-stranded DNA templates by the herpes simplex virus-induced DNA polymerase

The DNA polymerase encoded by herpes simplex virus 1 consists of a single polypeptide of Mr 136,000 that has both DNA polymerase and 3'----5' exonuclease activities; it lacks a 5'----3' exonuclease. The herpes polymerase is exceptionally slow in extending a synthetic DNA primer annealed to circular single-stranded DNA (turnover number approximately 0.25 nucleotide). Nevertheless, it is highly processive because of its extremely tight binding to a primer terminus (Kd less than 1 nM). The single-stranded DNA-binding protein from Escherichia coli greatly stimulates the rate (turnover number approximately 4.5 nucleotides) by facilitating the efficient binding to and extension of the DNA primers. Synchronous replication by the polymerase of primed single-stranded DNA circles coated with the single-stranded DNA-binding protein proceeds to the last nucleotide of available 5.4-kilobase template without dissociation, despite the 20-30 min required to replicate the circle. Upon completion of synthesis, the polymerase is slow in cycling to other primed single-stranded DNA circles. ATP (or dATP) is not required to initiate or sustain highly processive synthesis. The 3'----5' exonuclease associated with the herpes DNA polymerase binds a 3' terminus tightly (Km less than 50 nM) and is as sensitive as the polymerase activity to inhibition by phosphonoacetic acid (Ki approximately 4 microM), suggesting close communication between the polymerase and exonuclease sites.     

Characterization of 3''----5'' exonuclease associated with DNA polymerase of silkworm nuclear polyhedrosis virus.

3'----5' Exonuclease specific for single-stranded DNA copurified with DNA polymerase of nuclear polyhedrosis virus of silkworm Bombyx mori (BmNPV Pol). BmNPV Pol has no detectable 5'----3' exonuclease activity on single-stranded or duplex DNA. Analysis of the products of 3'----5' exonucleolytic reaction showed that deoxynucleoside monophosphates were released during the hydrolysis of single-stranded DNA. The exonuclease activity cosedimented with the polymerase activity during ultracentrifugation of BmNPV Pol in glycerol gradient. The polymerase and the exonuclease activities of BmNPV Pol were inactivated by heat with nearly identical kinetics. The mode of the hydrolysis of single-stranded DNA by BmNPV Pol-associated exonuclease was strictly distributive. The enzyme dissociated from single-stranded DNA after the release of a single dNMP and then reassociated with a next polynucleotide being degradated.     

Proofreading activity of DNA polymerase III responds like elongation activity to auxiliary subunits

A comparison of the 3'----5' proofreading properties between Escherichia coli DNA polymerase III holoenzyme and DNA polymerase III' was conducted. This study indicated that the influence of the holoenzyme auxiliary subunits on the proofreading exonuclease parallels their effect on the elongation reaction. At physiological ionic strengths the auxiliary subunits markedly stimulated the exonuclease rate in an ATP-dependent reaction, while the exonuclease rate of DNA polymerase III' was not affected by ATP. E. coli single-stranded DNA binding protein stimulated the 3'----5' exonuclease activity of holoenzyme and inhibited DNA polymerase III'. Similarly, the auxiliary subunits and ATP converted the proofreading activity to a highly processive exonuclease. Our observation, that the exonuclease activity of the DNA polymerase III holoenzyme responded to ATP, salt, and E. coli single-stranded DNA-binding protein like the elongation activity, is consistent with the polymerase and exonuclease subunits acting within the same complex in a coordinated reaction.     

Exonuclease activity associated with a multiprotein form of HeLa cell DNA polymerase alpha. Purification and properties of the exonuclease

The DNase that is associated with a multiprotein form of HeLa cell DNA polymerase alpha (polymerase alpha 2) has two distinct exonuclease activities: the major activity initiates hydrolysis from the 3' terminus and the other from the 5' terminus of single-stranded DNA. The two exonuclease activities show identical rates of thermal inactivation and coincidental migration during chromatofocusing, glycerol gradient centrifugation, and nondenaturing polyacrylamide gel electrophoresis of the DNase. Moreover, the purified DNase shows a single protein band of Mr 69,000 following nondenaturing polyacrylamide and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The 3'----5' exonuclease activity hydrolyzes only single-stranded DNA substrates and the products are 5' mononucleotides. This activity recognizes and excizes mismatched bases at the 3' terminus of double-stranded DNA substrates. The 3'----5' exonuclease does not hydrolyze 3' phosphoryl terminated single-stranded DNA substrates. The 5'----3' exonuclease activity also only hydrolyzes single-stranded DNA substrates. The rate of hydrolysis, however is only about 1/25th the rate of the 3'----5' exonuclease. This exonuclease activity requires a 5' single-stranded terminus in order to initiate hydrolysis and does not proceed into double-stranded regions. The products of hydrolysis by 5'----3' exonuclease are also 5' nucleoside monophosphates.     

Exonucleolytic proofreading enhances the fidelity of DNA synthesis by chick embryo DNA polymerase-gamma

The high fidelity of chick embryo DNA polymerase-gamma (pol-gamma) observed during in vitro DNA synthesis (Kunkel, T. A. (1985) J. Biol. Chem. 260, 12866-12874) has led us to examine this DNA polymerase for the presence of an exonuclease activity capable of proofreading errors. Highly purified chick embryo pol-gamma preparations do contain exonuclease activity capable of digesting radiolabeled DNA in a 3'----5' direction, releasing deoxynucleoside 5'-monophosphates. The polymerase and exonuclease activities cosediment during centrifugation in a glycerol gradient containing 0.5 M KCl. In the absence of dNTP substrates, this exonuclease excises both matched and mismatched primer termini, with a preference for mismatched bases. Excision is inhibited by the addition of nucleoside 5'-monophosphates to the digestion reaction. In the presence of dNTP substrates to permit competition between excision and polymerization from the mismatched primer, the exonuclease excises mismatched bases from preformed terminal mispairs with greater than 98% efficiency. The preference for excision over polymerization can be diminished by addition of either high concentrations of dNTP substrates or nucleoside 5'-monophosphates to the exonuclease/polymerase reaction. To determine if this exonuclease is capable of proofreading misinsertions produced during a normal polymerization reaction, a sensitive base substitution fidelity assay was developed based on reversion of an M13mp2 lacZ alpha nonsense codon. In this assay using reaction conditions that permit highly active exonucleolytic proofreading, pol-gamma exhibits a fidelity of less than one error for every 260,000 bases polymerized. As for terminal mismatch excision, fidelity is reduced by the addition to the synthesis reaction of high concentrations of dNTP substrates or nucleoside 5'-monophosphates, both hallmarks of exonucleolytic proofreading by prokaryotic enzymes. Taken together, these observations suggest that the 3'----5' exonuclease present in highly purified chick embryo pol-gamma preparations proofreads base substitution errors during DNA synthesis. It remains to be determined if the polymerase and exonuclease activities reside in the same or different polypeptides.     

The role of exonucleolytic processing and polymerase-DNA association in bypass of lesions during replication in vitro. Significance for SOS-targeted mutagenesis

The role of exonuclease activity in trans-lesion DNA replication with Escherichia coli DNA polymerase III holoenzyme was investigated. RecA protein inhibited the 3'----5' exonuclease activity of the polymerase 2-fold when assayed in the absence of replication and had no effect on turnover of dNTPs into dNMPs. In contrast, single-stranded DNA-binding protein, which had no effect on the exonuclease activity in the absence of replication, showed a pronounced 7-fold suppression of the 3'----5' exonuclease activity during replication. The excision of incorporated dNMP alpha S residues from DNA by the 3'----5' exonuclease activity of DNA polymerase III holoenzyme was inhibited 10-20-fold; still no increase in bypass of pyrimidine photodimers was observed. Thus, in agreement with our previous results in which the exonuclease activity was inhibited at the protein level (Livneh, Z. (1986) J. Biol. Chem. 261, 9526-9533), inhibition at the DNA level also did not increase bypass of photodimers. Fractionation of the replication mixture after termination of DNA synthesis on a Bio-Gel A-5m column under conditions which favor polymerase-DNA binding yielded a termination complex which could perform turnover of dNTPs into dNMPs. Adding challenge-primed single-stranded DNA to the complex yielded a burst of DNA synthesis which was promoted most likely by DNA polymerase III holoenzyme molecules transferred from the termination complex to the challenge DNA thus demonstrating the instability of the polymerase-DNA association. Addition of a fresh sample of DNA polymerase III holoenzyme to purified termination products, which consist primarily of partially replicated molecules with nascent chains terminated at UV lesions, did not result in any net DNA synthesis as expected. However, reactivation of lesion-terminated primers was achieved by pretreatment with a 3'----5' exonuclease which excised 200 nucleotides or more, generating new 3'-OH termini located away from the UV lesions. When these exonuclease-treated products were subjected to a second round of replication, an increased level of DNA synthesis was observed including additional bypass of photodimers. These results suggest the possibility that 3'----5' exonuclease processing might be required at least transiently during one of the stages of trans-lesion DNA replication, which is believed to be the mechanism of SOS-targeted mutagenesis.     

DNA polymerase gamma from Xenopus laevis. II. A 3'----5' exonuclease is tightly associated with the DNA polymerase activity

Xenopus laevis DNA polymerase gamma co-purifies with a tightly associated 3'----5' exonuclease. The purified enzyme lacks 5'----3' exonuclease and endonuclease activity. The ratio of the 3'----5' exonuclease activity to DNA polymerase gamma activity remains constant over the final three chromatographic procedures. In addition, these activities co-sediment under partially denaturing conditions in the presence of ethylene glycol. The associated 3'----5' exonuclease activity removes a terminally mismatched nucleotide more rapidly than a correctly base-paired 3'-terminal residue, as expected if this exonuclease has a proofreading function. The 3'----5' exonuclease has the ability to release a terminal phosphorothioated nucleotide, a property shared with T4 DNA polymerase, but not with Escherichia coli DNA polymerase I.     

Novel blunt-end addition reactions catalyzed by DNA polymerase I of Escherichia coli

DNA polymerase I (Klenow fragment) of Escherichia coli catalyzes the addition of deoxynucleotides to 3' hydroxyl termini of blunt-ended DNA fragments. The product of the reaction, which we call +1 addition, is found only in very low yield under conditions that permit editing by the 3'----5' exonuclease activity of the wild-type polymerase. A mutant form of the Klenow fragment that lacks detectable 3'----5' exonuclease activity shows an elevated accumulation of the +1 addition product. The mutant enzyme can use any one of the four dNTPs to carry out the reaction when each precursor is provided individually. However, in the presence of all four dNTPs the addition of dATP is strongly preferred. Suppression of the editing function of the wild-type polymerase through the use of high concentrations of exogenous deoxynucleoside monophosphates also results in a significant increase in the amount of +1 addition product formed. The presence of a high dNMP concentration also alters the specificity of the nucleotide addition reaction carried out by the wild-type enzyme. Thus, in addition to dATP, the dNTP which is complementary to the exogenous deoxynucleoside monophosphate, is also used in the +1 addition reaction. A similar effect of dNMPs on the specificity of nucleotide addition was obtained with the mutant Klenow fragment. These results define two pathways for the +1 addition reaction: one that does not require coding information from the DNA template and a second in which coding information is provided by the exogenous dNMP.     

Binding of captan to DNA polymerase I from Escherichia coli and the concomitant effect on 5'----3' exonuclease activity

Captan (N-[(trichloromethyl)thio]-4-cyclohexene-1,2-dicarboximide) was shown to bind to DNA polymerase I from Escherichia coli. The ratio of [14C] captan bound to DNA pol I was 1:1 as measured by filter binding studies and sucrose gradient analysis. Preincubation of enzyme with polynucleotide prevented the binding of captan, but preincubation of enzyme with dGTP did not. Conversely, when the enzyme was preincubated with captan, neither polynucleotide nor dGTP binding was blocked. The modification of the enzyme by captan was described by an irreversible second-order rate process with a rate of 68 +/- 0.7 M-1 s-1. The interaction of captan with DNA pol I altered each of the three catalytic functions. The 3'----5' exonuclease and polymerase activities were inhibited, and the 5'----3' exonuclease activity was enhanced. In order to study the 5'----3' exonuclease activity more closely, [3H]hpBR322 (DNA-[3H]RNA hybrid) was prepared from pBR322 plasmid DNA and used as a specific substrate for 5'----3' exonuclease activity. When either DNA pol I or polynucleotide was preincubated with 100 microM captan, 5'----3' exonuclease activity exhibited a doubling of reaction rate as compared to the untreated sample. When 100 microM captan was added to the reaction in progress, 5'----3' exonuclease activity was enhanced to 150% of the control value. Collectively, these data support the hypothesis that captan acts on DNA pol I by irreversibly binding in the template-primer binding site associated with polymerase and 3'----5' exonuclease activities. It is also shown that the chemical reaction between DNA pol I and a single captan molecule proceeds through a Michaelis complex.(ABSTRACT TRUNCATED AT 250 WORDS)     

Complete replication of templates by Escherichia coli DNA polymerase III holoenzyme

DNA polymerase III holoenzyme (holoenzyme) processively and rapidly replicates a primed single-stranded DNA circle to produce a duplex with an interruption in the synthetic strand. The precise nature of this discontinuity in the replicative form (RF II) and the influence of the 5' termini of the DNA and RNA primers were analyzed in this study. Virtually all (90%) of the RF II products primed by DNA were nicked structures sealable by Escherichia coli DNA ligase; in 10% of the products, replication proceeded one nucleotide beyond the 5' DNA terminus displacing (but not removing) the 5' terminal nucleotide. With RNA primers, replication generally went beyond the available single-stranded template. The 5' RNA terminus was displaced by 1-5 nucleotides in 85% of the products; a minority of products was nicked (9%) or had short gaps (6%). Termination of synthesis on a linear DNA template was usually (85%) one base shy of completion. Thus, replication by holoenzyme utilizes all, or nearly all, of the available template and shows no significant 5'----3' exonuclease action as observed in primer removal by the "nick-translation" activity of DNA polymerase I.     

Aphidicolin inhibits DNA polymerizing activity but not nucleolytic activity of Escherichia coli DNA polymerase II.

We have purified the DNA polymerase II of Escherichia coli from the recombinant strain carrying the plasmid which encodes the polB gene. We confirmed that the purified protein, of molecular weight 90,000, possesses a 3'----5' exonuclease activity in addition to DNA polymerizing activity in a single polypeptide. Its DNA polymerizing activity was sensitive to the drug aphidicoline, which is a specific and direct inhibitor of the alpha-like DNA polymerases including eukaryotic replicative DNA polymerases. Aphidicolin had no detectable effect on the 3'----5' exonuclease activity. The inhibition by aphidicolin on the polymerizing activity of polymerase II was competitive with respect to dNTP and uncompetitive with respect to template DNA. This mode of action is the same as that on eukaryotic DNA polymerase alpha. The apparent Ki value calculated from Lineweaver-Burk plots was 55.6 microM.     

A conserved 3'----5' exonuclease active site in prokaryotic and eukaryotic DNA polymerases

The 3'----5' exonuclease active site of E. coli DNA polymerase I is predicted to be conserved for both prokaryotic and eukaryotic DNA polymerases based on amino acid sequence homology. Three amino acid regions containing the critical residues in the E. coli DNA polymerase I involved in metal binding, single-stranded DNA binding, and catalysis of the exonuclease reaction are located in the amino-terminal half and in the same linear arrangement in several prokaryotic and eukaryotic DNA polymerases. Site-directed mutagenesis at the predicted exonuclease active site of the phi 29 DNA polymerase, a model enzyme for prokaryotic and eukaryotic alpha-like DNA polymerases, specifically inactivated the 3'----5' exonuclease activity of the enzyme. These results reflect a high evolutionary conservation of this catalytic domain. Based on structural and functional data, a modular organization of enzymatic activities in prokaryotic and eukaryotic DNA polymerases is also proposed.     

Coordinated leading- and lagging-strand synthesis at the Escherichia coli DNA replication fork. III. A polymerase-primase interaction governs primer size.

Studies with a rolling-circle DNA replication system reconstituted in vitro with a tailed form II DNA template, the DNA polymerase III holoenzyme (Pol III HE), the Escherichia coli single-stranded DNA binding protein, and the primosome, showed that within the context of a replication fork, the oligoribonucleotide primers that were formed were limited to a length in the range of 9 to 14 nucleotides, regardless of whether they were subsequently elongated by the lagging-strand DNA polymerase. This is in contrast to the 8-60-nucleotide-long primers synthesized by the primosome in the absence of DNA replication on a bacteriophage phi X174 DNA template, although when primer synthesis and DNA replication were catalyzed concurrently in this system, the extent of RNA polymerization decreased. As described in this report, we therefore examined the effect of the DNA Pol III HE on the length of primers synthesized by primase in vitro in the absence of DNA replication. When primer synthesis was catalyzed either: i) by the primosome on a phi X174 DNA template, ii) by primase on naked DNA with the aid of the DnaB protein (general priming), or iii) by primase alone at the bacteriophage G4 origin, the presence of the DNA Pol III HE in the reaction mixtures resulted in a universal reduction in the length of the heterogeneous RNA products to a uniform size of approximately 10 nucleotides. dNTPs were not required, and the addition of dGMP, an inhibitor of the 3'----5' exonuclease of the DNA Pol III HE, did not alter the effect; therefore, neither the 5'----3' DNA polymerase activity nor the 3'----5' exonuclease activity of the DNA Pol III HE was involved. E. coli DNA polymerase I, and the DNA polymerases of bacteriophages T4 and T7 could not substitute for the DNA Pol III HE. The Pol III core plays a crucial role in mediating this effect, although other subunits of the DNA Pol III HE are also required. These observations suggest that the association of primase with the DNA Pol III HE during primer synthesis regulates its catalytic activity and that this regulatory interaction occurs independently of, and prior to, formation of a preinitiation complex of the DNA Pol III HE on the primer terminus.     

Complete enzymatic synthesis of DNA containing the SV40 origin of replication

The replication of simian virus 40 origin-containing DNA has been reconstituted in vitro with SV40 large T antigen and purified proteins isolated from HeLa cells. Covalently closed circular DNA (RF I') daughter molecules are formed in the presence of T antigen, a single-stranded DNA binding protein and DNA polymerase alpha-primase complex, together with ribonuclease H, DNA ligase, topoisomerase II, and a double-stranded specific exonuclease that has been purified to homogeneity. The 44-kDa exonuclease-digested oligo(rA) annealed to poly(dT) in the 5'----3' direction. DNA ligase and the 5'----3' exonuclease were essential for RF I' formation. Covalently closed circular duplex DNA and full length linear single-stranded DNA were detected by alkaline gel electrophoresis as products of the complete system. DNA replication in the absence of either DNA ligase or the 5'----3' exonuclease yielded DNA products that were half length (approximately 1500 nucleotides) and smaller Okazaki-like fragments (approximately 200 nucleotides). Hybridization experiments showed that the longer chains were synthesized from the leading strand template, while the small products were synthesized from the lagging strand template. These results suggest that the RNA primers attached to 5' ends of replicated DNA are completely removed by the 5'----3' exonuclease, with the assistance of RNase H.     

Alteration of the exonuclease activities of DNA polymerase I by captan

DNA polymerase I is a multifaceted enzyme with one polymerizing and two exonuclease activities. Captan was previously shown to be an inhibitor of this enzyme's polymerizing activity and this report measures the effects of captan on the two exonuclease activities. When the holoenzyme was tested, captan enhanced the degradation of poly(dA-dT), T7 DNA and, to a significantly lesser extent, heat-denatured DNA. However, when the effects of captan were tested as a function of substrate concentration, the stimulatory influence was measured only at high substrate concentrations. At low concentrations of DNA, captan was inhibitory. Inhibition and enhancement each showed an ED50 of the same value (approx. 100 microM). By assaying the two exonuclease activities separately it was shown that the differential effect on the holoenzyme by captan was the result of a combined inhibition of the 3'----5' exonuclease and enhancement of the 5'----3' exonuclease. Klenow fragment with poly(dA-dT) as substrate was used to assay for 3'----5' exonuclease activity. Captan inhibited this exonuclease and the inhibition could be prevented by the addition of greater concentrations of substrate. Holoenzyme and poly(rA)-poly(dT) were used to assay for 5'----3' exonucleolysis, which was enhanced at higher concentrations of substrate in the presence of captan.     

Modulation of hyperthermophilic DNA polymerase activity by archaeal chromatin proteins

Sulfolobus synthesizes a large quantity of highly conserved 7-kDa DNA-binding proteins suspected to be involved in chromosomal organization. The effect of the 7-kDa proteins on the polymerization and 3'-5' exonuclease activities of a family B DNA polymerase (polB1) from the hyperthermophilic archaeon Sulfolobus solfataricus was investigated. polB1 degraded both single-stranded DNA and double-stranded DNA at similar rates in vitro at temperatures of physiological relevance. The 7-kDa proteins were capable of significantly inhibiting the excision and enhancing the extension of matched template primers by the polymerase. However, the proteins did not protect single-stranded DNA from cleavage by polB1. In addition, the 7-kDa proteins did not affect the proofreading ability of polB1 and were not inhibitory to the excision of mismatched primers by the polymerase. The dNTP concentrations required for the effective inhibition of the 3'-5' exonuclease activity of polB1 were lowered from approximately 1 mm in the absence of the 7-kDa proteins to approximately 50 microm in the presence of the proteins at 65 degrees C. Our data suggest that the 7-kDa chromatin proteins serve to modulate the extension and excision activities of the hyperthermophilic DNA polymerase, reducing the cost of proofreading by the enzyme at high temperature.     

Coprecipitation of 3'----5' exonuclease and DNA polymerase gamma with anti-DNA polymerase gamma antibody

Highly purified preparations of chick embryo DNA polymerase gamma contained 3'----5' exonuclease activity which might be responsible for the exonucleolytic proofreading during DNA synthesis [Kunkel, T.A. & Soni, A. (1988) J. Biol. Chem. 262, 4450-4459]. A rabbit antibody produced against highly purified chick DNA polymerase gamma precipitated 3'----5' exonuclease activity to the same extent as DNA polymerase gamma activity. Furthermore, the antibody neutralized the two enzyme activities to an equal extent. However, the exonuclease activity was more resistant than DNA polymerase gamma activity to thermal treatment at 50 degrees C, although both activities were partially protected with polynucleotides. The results obtained suggest that these two enzymes are associated as a single enzyme complex or that the two activities reside in a single molecule, and the active site of DNA polymerase gamma and 3'----5' exonuclease are, although not identical, closely correlated.     

Inactivation of DNA polymerase I (Klenow fragment) by adenosine 2',3'-epoxide 5'-triphosphate: evidence for the formation of a tight-binding inhibitor

The suicidal inactivation of Escherichia coli DNA polymerase I by epoxy-ATP has been previously reported (Abboud et al., 1978). We have examined in detail the mechanism of this inactivation utilizing a synthetic DNA template-primer of defined sequence. Epoxy-ATP inactivates the large fragment of DNA polymerase I (the Klenow fragment) in a time- and concentration-dependent manner (KI = 21 microM; kinact = 0.021 s-1). Concomitant with inactivation is the incorporation of epoxy-AMP into the primer strand. The elongated DNA duplex directly inhibits the polymerase activity of the enzyme (no time dependence) and is resistant to degradation by the 3'----5' exonuclease and pyrophosphorylase activities of the enzyme. Inactivation of the enzyme results from slow (4 X 10(-4) s-1) dissociation of the intact epoxy-terminated template-primer from the enzyme and is thus characterized as a tight-binding inhibition. Surprisingly, while the polymerase activity of the enzyme is completely suppressed by epoxy-ATP, the 3'----5' exonuclease activity remains intact. The data presented demonstrate that even though the polymerase site is occupied with duplex DNA, the enzyme can bind a second DNA duplex and carry out exonucleolytic cleavage.     

Hydrolysis of 3'-terminal mispairs in vitro by the 3'----5' exonuclease of DNA polymerase delta permits subsequent extension by DNA polymerase alpha

Purified DNA polymerase alpha, the major replicating enzyme found in mammalian cells, lacks an associated 3'----5' proofreading exonuclease that, in bacteria, contributes significantly to the accuracy of DNA replication. Calf thymus DNA polymerase alpha cannot remove mispaired 3'-termini, nor can it extend them efficiently. We designed a biochemical assay to search in cell extracts for a putative proofreading exonuclease that might function in concert with DNA polymerase alpha in vivo but dissociates from it during purification. Using this assay, we purified a 3'----5' exonuclease from calf thymus that preferentially hydrolyzes mispaired 3'-termini, permitting subsequent extension of the correctly paired 3'-terminus by DNA polymerase alpha. This exonuclease copurifies with a DNA polymerase activity that is biochemically distinct from DNA polymerase alpha and exhibits characteristics described for a second replicative DNA polymerase, DNA polymerase delta. In related studies, we showed that the 3'----5' exonuclease of authentic DNA polymerase delta, like the purified exonuclease, removes terminal mispairs, allowing extension by DNA polymerase alpha. These data suggest that a single proofreading exonuclease could be shared by DNA polymerases alpha and delta, functioning at the site of DNA replication in mammalian cells.