Mutation Research, 250 ( 1991 ) 175-182
175
© 1991 Elsevier Science Publishers B.V. All rights reserved IX127-5107/91/$03.50 ADONIS 002751079100176F
MUT 02510
The fidelity of DNA synthesis by the catalytic subunit of yeast DNA polymerase a alone and with accessory proteins Thomas A. Kunkel, John D. Roberts and Akio Sugino Laboratory of Molecular Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 (U.S.A.) (Accepted 5 April 1991)
Keywords: Fidelity; DNA polymerase a; DNA replication
Summary The fidelity of DNA synthesis catalyzed by the 180-kDa catalytic subunit (p180) of DNA polymerase a from Saccharomyces cerecisiae has been determined. Despite the presence of a 3 ' ~ 5' exonuclease activity (Brooke et al., 1991, J. Biol. Chem., 266, 3005-3015), its accuracy is similar to several exonuclease-deficient DNA polymerases and much lower than other DNA polymerases that have associated exonucleolytic proofreading activity. Average error rates are 1/9900 and 1/12000, respectively, for single base-substitution and minus-one nucleotide frameshift errors; the polymerase generates deletions as well. Similar error rates are observed with reactions containing the 180-kDa subunit plus an 86-kDa subunit (p86), or with these two polypeptides plus two additional subunits (p58 and p49) comprising the DNA primase activity required for DNA replication. Finally, addition of yeast replication factor-A (RF-A), a protein preparation that stimulates DNA synthesis and has single-stranded DNA-binding activity, yields a polymerization reaction with 7 polypeptides required for replication, yet fidelity remains low relative to error rates for semiconservative replication. The data suggest that neither exonucleolytic proofreading activity, the /3 subunit, the DNA primase subunits nor RF-A contributes substantially to base substitution or frameshifl error discrimination by the DNA polymerase a catalytic subunit.
DNA replication in extracts of human (Roberts and Kunkel, 1988) or monkey (Hauser et al., 1988) cells is highly accurate. In human cell extracts, this accuracy results both from high nucleotide selectivity and from exonucleolytic proofreading on both the leading and the lagging strand (Roberts et al., 1991). One approach for under-
Correspondence: Dr. Thomas A. Kunkel, Laboratory of Molecular Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 (U.S.A.), Phone: (919)-541-2644; Fax: (919)-541-7593.
standing the molecular details of mutational processes during replication is to define the proteins responsible for these two discrimination steps. The observations that lagging strand replication is accurate and that exonucleolytic proofreading occurs on the lagging strand are striking when considered in the context of the current model for the eukaryotic replication fork. This model, which is supported by a large body of evidence in vitro (e.g., Kelly, 1988; Tsurimoto et al., 1990; Hurwitz et al., 1990), posits that DNA polymerase a is responsible for lagging strand synthesis. However, mammalian DNA poly-
176 mcrase c~ preparations have historically (for rcview, see Fry. and I_,mb, 1986) bccn found Io be devoid of 3 ' ~ 5' exonuclease activity, although exceptions have been noted (see below}. Furthermore. based on our experience with preparations from a variety of sources (Kunkel, 1985: Roberts and Kunkel, 1988: Kunkel et al.. 1989), the fidelity of D N A synthesis by D N A polymcrasc (~ is consistently much lower than that of lagging strand replication (Roberts et al., 1991 ). Given these observations, it is reasonable to ask whether more accurate forms of D N A polymerase ~e can bc found, tligher accuracy might result if the polymerasc has an associated 3' --* 5' exonucleasc a n d / o r additional associated proteins that could enhance nucleotide selectivity. An ideal place to begin thc scarch for such fidelity factors is in Saccharomvces cererisiue. This organism not only provides a powerful genetic system, but is similar to mammalian systcms regarding its replication enzymology (e.g., Burgers. 1989: Brill and Stillman, 1989). This similarity includes the involvement of D N A polymerasc ~ (formerly designated D N A poIymerase I) as an esscntial rcplicative polymerasc. The catalytic subunit of yeast I)NA polymerase Ce, like its mammalian counterpart, is a 180-kDa polypeptide (Plevani et al., 1985). When overproduccd in yeast from the cloned gene and then extensively purified (Brooke et al., 1991), this plS0 protein not only has polymerization activity but also 3'--~ 5' exonucleolytic activity, effectively digesting a hmg, synthetic, singlestranded po[ynucleotide substrate. This exonuc[ease is thus a candidate for a proofreading activity. Similar to its mammalian equivalent, yeast I)NA polymerase ~r can be purified by immunoaffinity chromatography as a four-subunit complex (for revicw, see Burgers, 1989). In addition to the catalytic p180, this complex contains an 86-kDa polypeptide (p86) that has no apparent enzymatic activity but stabilizes the polymerase activity against thermal inactivation (Brooke ct al., 1991). The g e n t coding for p86 has also been cloned, allowing overproduction of the protein in yeast and subsequent extensivc purification (Brooke ct al., 1991). The D N A polymerasc a complex also contains two polypeptides (58- and 49-kDa, designated p58 and p49, respec-
lively) that together comprise a DNA primasc activity for initiating lagging strand ()kazaki fragnlents.
The similarity of the yeast and mammalian replication systems extends to other proteins required for scmiconservative I)NA replication. Among these is a yeast single-stranded I)NAbinding protein that can participate in origin-dependent initiation (Brill and Stillman, 1989). The preparation actually consists of three polypcptides (69, 3¢~ and 13 kDa), collectively referred to as replication factor-A (RF-A). In addition to its role in intiation, RF-A is thought to participate in chain elongation reactions on I~oth the leading and lagging strands. As prokarvotic singlestranded I)NA binding proteins have been shown to intluence the fidelity of DNA synthesis with purified DNA polymerases (Kunkcl et al.. 1979L it was of interest to examine the influence of yeast RF-A on the fidelity of yeast I)NA polymeras¢ ¢~. In the present study, wc have used assays that monitor base-substitution and frameshift error rates during a single cycle of DNA synthesis in vitro (Kunkel ct al., 1989), to examine the fidelity of piN0 alone or in the presence of pN6, p86 plus p58 and p49, and p86, p58, p49 and R[:-A. The data suggest that, despite the presence of a 3' --, 5' exonuclease activity and six additional proteins, DNA synthesis catalyzed by the catalytic subunit of yeast DNA polymerase ~ is not highly accurate, being much less accurate than lagging strand I)NA replication in cxtracts of human cells. Materials and methods
Materials The bacterial strains, bacteriophage M l 3 m p 2 and its derivatives and other reagents required for the fidelity assays have been described (Kunkel et al,, 1989 and refs. therein). The p180 and p86 proteins were kindly provided by G. Brooke (University of Illinois at Champaign-Urbana). These were overproduced, purified and characterized as described (Brooke el al., 1991). DNA polymerase-¢~-DNA primase was immunoaffinitypurified as described (Kunkcl et al., 1989). The purification and characterization of yeast RF-A will be described elsewhere (A. Sugino, in prepa-
177
ration). Its properties and polypeptide compostion are similar to those described by Brill and Stillman (1989).
Methods Ml3mp2 mutagenesis assays. Three assays for measuring the fidelity of a single round of gapfilling DNA synthesis were employed. In all three cases, a gapped, double-stranded Ml3mp2 DNA substrate is constructed (Kunkel and Soni, 1988) such that the single-stranded gap contains the mutational target. (i) Forward-mutation assay. The forward-mutation assay (Kunkel, 1985) uses thc wild-type lacZa sequence. Correct polymerization to fill the 390-base gap produces DNA that, when used to transfect an E. coli host strain, will produce dark blue M13 plaques. Errors during synthesis are scored as lighter blue or colorless plaques. Since the assay measures loss of a gene function (a-complementation of /3-galactosidasc activity) that is not essential for phage production, a wide variety of mutations at many different sites can be recovered and scorcd. This includes 227 differcnt single base substitution errors at 116 different template positions, single-base frameshifts at 150 different template positions involving both nonreiterated and reiterated bases, deletions and complex errors. (ii) Opal codon reversion assay. It is possible to focus exclusivcly on single-base substitution errors using the opal codon reversion assay (Kunkel and Soni, 1988). In this assay thc 361-base gap contains a single base change [G ~ A in the viral (plus) template strand at position 89 of the lacZa coding sequence]. This change creates an opal (TGA) codon, resulting in a colorless-plaque phenotype. Base-substitution errors resulting from gap-filling synthesis are detected as blue plaques and the reversion frequency (the proportion of blue to total plaques) reflects the base substitution error rate. 8 of 9 possible basc-substitution errors at the TGA opal codon yield a dctectable bluc plaque phenotype. (iii) Assay for minus-one base frameshifts. This assay (Bebenek and Kunkel, 1990) also employs an M13mp2 DNA substrate containing a 361-base gap and also scores blue revertants of a colorless
plaque phenotype. In this instance one begins with a mutant that is colorless due to the addition of one extra base in the lacZa coding sequence, a T into the run of four consecutive Ts at positions 70-73 (using our previous convention (22) this is referred to as the + T70 mutation). Blue plaques result from errors that restore the reading frame.
p180-p86 complex formation. 13 ng each of p180 and p86 were incubated for 2 h at 0 ° C in 10 mM Hepes (pH 7.4), 2.5 mM dithiothreitol. This has been shown to yield complex formation as determined by co-immunoprecipitation of the two polypeptides (Brooke et al., 1991). DNA polymerase reactions. Reactions (50 gl) contained 20 mM Hepes (pH 7.4), 2 mM dithiothreitol, 6 mM MgCI 2, 6.5% (for p180 or polymerase-primase) or 13% (p180-p86) glycerol, 10 /zg BSA, 150 ng gapped M13mp2 DNA (either wild-type or mutant, depending on the assay), either 13 ng of p180, 26 ng of p180-p86 complex or 5 units of polymerase-primase complex, and, unless noted otherwise, 500 /zM each of dATP, dTI'P, d G T P and dCTP. After incubation at 30 ° C for 2 h, reactions were terminated by addition of E D T A to a final concentration of 1(1 mM. 15-p.l aliquots of each reaction were analyzed by agarose gel electrophoresis as described (Kunkel, 1985). For all conditions, the gap was filled to the extent that the DNA migrated coincident with a fully double-stranded, nicked circular (RFII) DNA standard (not shown). Other procedures. Preparation of M13mp2 DNA substrates, transfection of competent ceils, plating and scoring of o~-complcmentation mutants and DNA sequence analysis were performed as described (Kunkel et al., 1989 and refs. therein). Results
We recently demonstrated (Kunkel et al., 1989) that yeast DNA polymerase o~, whether purified by conventional chromatography or by immunoaffinity chromatography as a polymeraseprimase complex, did not excise a 3'-OH-terminally mispaired nucleotide from either of two
178
different DNA substrates, even after extended incubation with a substantial excess of polymerase. This lack of exonucIeolytic activity was confirmed by Brooke et al. (1991), who demonstrated that plS0, with or without the additional polypeptides of the polymerase-primase complex, failed to degrade the radiolabeled oligo (dC)l.~ mispaired portion of an oligo(dA)ts(dC)~ -poly (dT) substrate, under conditions in which several DNA polymerases containing proofreading activity readily digested this substrate. However, Brooke et al. (1991) also found that all three forms of yeast DNA polymerase o~ (plS(I, p180-p86 and the polymcrase-primase complex) released dCMP from the 3'-OH terminus of a single-stranded s u b s t r a t c , poly(dT)~,0~j-32 PdCMP0. 4. The exonuclease and polymerasc activities co-immunoprecipitated using antibodies directed against either p18(1 or p86, suggesting an association between the two activities. The detection of this polymcrase-associated 3' ~ 5' cxonuclease activity prompted us to examinc the fidelity of all three forms of yeast DNA polymerase ot under conditions where proofreading activity might be expected to yield high fidelity DNA synthesis. For this purpose we used the opal codon reversion assay (sec Methods) for base-substitution errors that arc known to be proofread effectively by a variety of DNA polymerases having associated exonuclease activity. Gap-filling synthesis reactions contained relatively low concentrations of
dNTPs, in order to detect any proofreading activity that might be present. As expected, DNA synthesis by enzymes having associated proofreading activity was highly accuratc. Low reversion frequencies were observed (Table 11 for the large fragment of E. coli DNA polymerase i (Bebenek and Kunkel, 1990), bovine DNA polymerase • (formerly designated DNA polymerase 611, Kunkel et al., 1987), chick-embryo DNA polymcrase y (Kunkel and Soni, 1988), porcineliver DNA polymerase y (Kunkel and Mosbaugh, 1989) and T7 DNA polymerase (data not shown). In contrast, the catalytic subunit of yeast DNA polymcrase o~ generated revertants at a frequency that was - 100-fold above the background frcquency of uncopied DNA. Similar fidelity was observed when p86 was present or when the intact, immunoaffinity-purified DNA polymerasc ~e-DNA primasc complex (Table 1 and Kunkel ct al., 19891 was employed. The fidelity of these forms of yeast pol o~ was comparable to that of the cxonuclease-deficicnt form of the large fragment of E. coli DNA polymerase 1 (Bcbcnek and Kunkel, 199(I, and Tablc 1). The error ratc per dctectable nucleotidc polymerized, ~ 1/100(10, is comparable to the estimated base selectivity of a variety of exonuclcasc-deficient DNA polymcrascs (for review, scc Echols and Goodman, 1991). The data suggest that the exonuclease activity detected that releases dCMP from singlestranded poly(dT)~,~-~-'P-dCMP~,, does not contribute significantly to the base-substitution fi-
TABLE 1 Base-substitution fidelity of yeast plS0, p180-p86 and polymerase-primasc compared to that of DNA Ix)lymerases known t¢) proofread errors DNA polymerase
3' --, 5'
Plaques scored
used
exonuclease activity
Total
Mutant
frequency ( × 10 - ~')
F. coli Pol I, large fragment bovine DNA polymerase ~ chick DNA polymerase y porcine DNA polymerase y yeast p180 yeast p 180 + p86 yeast polymerase-primase E. coli Pol l, large fragment
+ + + +
3 700(l~10 2 3000(,Ill 600000 6 100000 1100(X) 14(}000 730000 460{XXI
69 12 5 32 21 15 78 56
18 5.2 8.3 5.2 190 1 Ill 110 120
-
The background mutant frequency of uncopied DNA was 2 × 10 -~'.
Reversion
179 TABLE 2 Fidelity of yeast p180, p l S 0 - p 8 6 and polymerase-primase in the forward mutation assay D N A polymerase
Plaques scored
used
Total
Mutant frequency ( × 111-4)
Mutant
Expt. I p180 p 180 + p86 Polymerase-primase (Kunkel et al., 1989)
2591 4 ill 3
55 92
6 328 6 284 6 409
279 296 270
210 230 140 +_20
Expt. 2 p 180 p 180 + p86 Polymerase-primase
440 47(1 420
The background mutant frequency was 6 . 7 x 10 4
delity of gap-filling synthesis by yeast DNA polymerase a under the reaction conditions employed in this study. The opal codon reversion assay monitors single base-substitution errors at three template nucleotide positions. Since DNA polymerase error rates have been found to depend on the local sequence around the error as well as on the composition and symmetry of the error (for review, see Kunkel and Bebenek, 1988), we next examined the fidelity of the three yeast pol a forms in the M13mp2 forward mutation assay to see if differences could be found. This assay scores a wide variety of mutations at many different sites (see Methods). As for the reversion assay, all three forms of polymerase had comparable accuracy, in two independent experiments (Table 2). Again, the forward mutant frequency values are comparable to those generated by several DNA polymerases that lack proofreading activity (Kunkel, 1985; Roberts and Kunkel, 1988; Kunkel et al., 1989). DNA sequence analysis of 134 mutants generated by p180 and of 20 mutants generated in reactions containing p180 plus p86 (Table 3) demonstrate that both forms of polymerase generate single base-substitutions, single-base frameshifts and deletions of larger numbers of nucleotides, that the two forms of polymerase have comparable accuracy, and that their error rates and error specificities are remarkably similar to those of the DNA polymerase a-primase complex described earlier (Kunkel et al., 1989).
In considering which accessory proteins might contribute to the high fidelity of semiconservative replication, an attractive candidate is yeast RF-A. This three-subunit yeast protein preparation stimulates synthesis by yeast DNA polymerase a (not shown) and the largest subunit (69 kDa) has single-stranded DNA-binding activity (Brill and Stillman, 1989). In addition to its suggested role in initiation at a replication origin (Brill and Stillman, 1989), yRF-A has been suggested to participate in the elongation phase of the reaction. We therefore asked if addition of yRF-A to reactions catalyzed by the presumed replicative form of yeast pol a, the DNA polymerase otprimase complex, altered fidelity. The addition of yRF-A had no apparent effect
TABLE 3 Types of errors as defined by D N A sequence analysis Type of
p180
mutation
NumM.F. NumM.F. b e r o f ( x l 0 -4) b e r o f (xl0 mutant mutants
B a ~ substititions 44 O n e base frameshifts 48 Deletions > 1 base 31 Other a 11 Total 134
plS0-p86
69 75 49 17 210
7 7 3 3 20
4)
80 80 35 35 230
a This class includes mutants containing more than a single nucleotide difference from wild-type or no changes within the lacZa sequence present within the gapped D N A substrate.
180 TABI ,E 4 Fidelity of yeast [)NA polymcrasc (r-primase with and without RF-A Components
Plaques scored
used
Total
Mutant
Mtlt[mt frequency { × I0
a)
E3cpt, 1. Forward Mutation Assa.v Polymerase- primasc Polymcrase. primase plus yRF-A
6 572 5 167
t)l 79
1411 1511
9901Xl
1112 1211
lilO I(H)
l'[xpt. 2. Mimt.s-one Nlwleotlde Ret'erstotl .,'|.s.';ay Polymerase -primase Polymerase -primase plus yRF-A
12()01X)
The background mutant frequency for the forward assay was 6.7 × 10 4 and for the reversion assay was 2,4 × l(I 5 For reactions containing yRF-A, 2 # g was added.
in the forward mutation assay (Table 4, Expt. 1). Among the mutants recovered, the ratio of light blue to colorless mutants was similar, providing suggestive evidence that yRF-A was not strongly influencing error specificity. Rather than sequencing another large collection of mutants to examine this further, wc determined the effect of addition of yRF-A on the frameshift fidelity of the polymerase-primase. Using a reversion assay (Bebenek et al., 1990) that scores minus-onenuclcotide errors at a template T V I T I ' sequencc (as dark blue revertants) or at 38 non-reiterated template nucleotide positions (as lighter blue revertants), the addition of yRF-A was found to have no effect on thc frameshift fidelity of thc polymerase oe-primase complex (Table 4, Expt. 2). Discussion The major conclusion from the present study is that the average fidelity of DNA synthesis catalyzed by replicative yeast DNA polymerase a is low relative to the estimated accuracy of semiconservative replication, despite the prescncc of 6 additional replication proteins and the reported presence of an associated 3'--*5' exonuclease activity. These data suggest that the primary determinants of selectivity against base-substitution and frameshift errors arc present in the p180 catalytic subunit itself, and that, at least in the in vitro system used here, the exonuclease is not contributing significantly to the fidelity of DNA polymcrasc oe. The latter conclusion is consistent
with the known DNA substratc requirement fl)r the 3 ' - - , 5 ' exonucleasc activity in the plSI) preparation. Thc exonuclcase only degraded a long singlc-strandcd polynucleotide but did not excise a terminal mismatch from a primer-template (Brooke et al., 1991), as has bccn observcd with 3' ---, 5' exonucleascs that proofread errors. The lack of an effect on fidelity with p86 and the primase subunits is consistent with numerous reports that these subunits of DNA polymerase ce-primasc complexes have littlc effect on other properties of the polymerization reaction (Wong et al., 1986; Cottcrill et al., 1987; Brooke el al.. 1991). Givcn the ability of prokaryotic singlestranded binding proteins to altcr the base-substitution error rates of DNA polymerases in vitro (Kunkel et al., 1979), the lack of an cffcct by yRF-A in either the forward mutation assay or the frameshift reversion assay was somewhat unexpected. At least for minus-one nucleotide frameshift errors at the T T I T F run (Table 4, Expt. 2), this lack of an effcct is not unreasonable. Such errors probably result from templateprimer misalignments whcrein the extra template base would be present in an otherwise doublcstranded primer stem (c.g., see Fig. 3 in Kunkel. 1990) and therefore unavailable to the singlestranded binding protein. The absence of an yRF-A effect on polymerasc fidelity may reflect the possibility that it could serve a diffcrent role in replication than has been attributed to prokaryotic SSBs. For example, its role may be limited to initiation of replication at the origin,
181
while another SSB participates in the chain elongation phase of replication. It is of course possible that the absence of fidelity-enhancing effects may reflect inappropriate reaction conditions or less than ideal protein-protein interactions. Thus, these proteins could still be important fidelity components in the context of a more complete replication apparatus. It is also possible that these proteins could improve fidelity for certain types of errors (by position or composition) that have not yet been exhaustively examined in the present study. For example, it is not unreasonable to expect that a protein like yRF-A that can bind to singlestranded DNA might reduce the frequency of deletion errors between directly repeated DNA sequences that are known to result from polymerization by DNA polymerase a (Table 3 and Kunkel et al., 1989). There is in fact precedent for this; when a 38-kDa SSB from yeast (one now known to probably be unimportant for DNA replication) was included in reactions catalyzed by yeast DNA polymrase a, fidelity was increased for deletions between direct repeats (Roberts et al., 19901. This possibility could be tested with yRF-A by sequence analysis of a large collection of mutants generated in the forward mutation assay. The fidelity studies reported here may have relevance beyond understanding error discrimination during semiconservative DNA replication. Substantial evidence supports a role for DNA polymerase a in repair of mismatched bases (Holmes et al., 1990; Thomas et al., 1991) and DNA damage (for review, see Fry and Loeb, 1986). The subunit composition of DNA polymerase a serving these roles may well be different. For example, the primase subunits may not be associated with the polymerase catalytic subunit. In the absence of additional information, the fidelity of the 180-kDa catalytic subunit alone reported here could be considered a baseline for comparison to future studies of the fidelity of DNA polymerase a-dependent repair synthesis.
Acknowledgements We would like to thank Glenn Brooke for generously providing p180 and p86 proteins and
Kenneth Tindall and James Clark for critical evaluation of the manuscript.
References Bebenek, K., C.M. Joyce, M.P. Fitzgerald and T.A. Kunkel (19901 The fidelity of DNA synthesis catalyzed by derivatives of Escherichia coli DNA polymerase I, J. Biol. Chem., 265, 13878-13887. Brill. S.J., and B. Stillman 119891 Yeast replication factor-A functions in the unwinding of the SV40 origin of DNA replication, Nature (lamdon), 342, 92-95. Brooke, R.G., R. Singhal, D.C. Hinkle and L.B. Dumas 119911 Purification and characterization of the 1811- and 86-kilodalton subunits of the Saccharomyces cereHsiae DNA primase-DNA polymerase protein complex, The 180 kilodalton subunit has both DNA polymerase and 3' --* 5' exonuclease activities, J. Biol. Chem., 266, 3005-31115. Burgers, P.J. 11989) Eukaryotic DNA polymerases a and h: Conserved properties and interactions, from yeast to mammalian cells, Progress in Nucleic Research and Molecular Biology, 37, 235-280. Cotterill, S.M., M.E. Reyland, L.A. Loeb and I.R. Lehman (19871 A cryptic proofreading 3' --, 5' exonuclease associated with the polymerase subunit of the DNA I~lymeraseprimase from Drosophila melanogaster, Proc. Natl. Acad. Sci. (U.S.A.). 84, 5635-5639. Echols, H., and M.F. Goodman (1991) Fidelity mechanisms in DNA replication, Annu. Rev. Biochem., in press. Fry, M., and L.A. Loeb (1986) in: Animal cell DNA polymerases. CRC Press, Inc., Boca Raton, FL. Hauser, J., A.S. Levine and K. Dixon (1988) Fidelity of DNA synthesis in a mammalian in vitro replication system, Mol. Cell. Biol., 8. 3267-3271. Holmes Jr., J., S. Clark and P. Modrich (1990) Strand-specific mismatch correction in nuclear extracts of human and Drosophila melanogaster cell lines, Proc. Natl. Acad. Sci. (U.S.A.), 87, 5837-5841. tlurwitz, J., F.B. Dean, A.D. Kwong and S.-ll. Lee (1990) The in vitro replication of DNA containing the SV4(1 origin, J. Biol. Chem., 265, 18043-18046. Kelly, T.J. (1988) SV40 DNA replication, J. Biol. Chem., 263, 17889-17892. Kunkel, T.A. (19851 The mutational specificity of DNA I:x~lymerase-/3 during in vitro DNA synthesis, Production of frameshift, base substitution and deletion mutations. J. Biol. Chem., 260, 5787-5796. Kunkel. T.A., and K. Bebenek (1988) Recent studies of the fidelity of DNA synthesis, Biochim. Biophys. Acta. 951, 1-15. Kunkel, T.A., and D.W. Mosbaugh (1989) Exonucleolytic proofreading by a mammalian DNA polymerase gamma, Biochemistry, 28, 988-995. Kunkel, T.A., and A. Soni (19881 Exonucleolytic proofreading enhances the fidelity of DNA synthesis by chick embryo DNA polymerase-y, J. Biol. Chem., 263, 4450-4459.
182 Kunkel, T.A., R.R. Meyer and L.A. Lx~eb(1979) Single-strand binding protein enhances fidelity of DNA synthesis in vitro, Proc. Natl. Acad. Sci. (U.S.A.). 76, 6331-6335. Kunkel, T.A.. R.D. Sabatino and R.A. Bambara (19871 Exonucleolytic proofreading by calf thymus I)NA polymerase 6, Proc. Natl. Acad. Sci. (U.S.A.), 84. 4865 4869. Kunkel, T.A., R.K. Hamatakc, J. Motto-Fox, M.P. Fitzgerald and A. Sugino (19891 The fidelity of DNA polymerasc I and the DNA polymerase I-DNA primasc complex from Saccharomyces cererisiae, Mol. Ceil. Biol.. 9, 4447-4458. Plcvani, P., M. Foiani, P. Valsasnini, G. Badaracco, IL Cheriathundam, and I,.M.S. Chang (19851 Polypcptidc structure of DNA primase from a yeast DNA polymerase primase complex, J. Biol. ('hem.. 26(I, 71112.-71117. Roberts, J.D., and T.A. Kunkcl (1988) Fidelity of a human cell DNA replication complex, Prtvc. Natl. Acad. Sci. (U.S.A.), 85, 7064-7068. Roberts, J.D., R.K. Hamatakc, M.S. Fitzgerald, A. Sugino, and T.A. Kunkel (19901 Effect of accessory' proteins on the fidelity of DNA synthesis by eukaryotic rcplicative poly-
merases, in M.L. Mcndelsohn and R.J. Aibertini (l{ds.). Progress in Clinical and Biological Researcll. Vol. 34{)A, Mutation and the Environment, Part A: Basic Mechanisms, Wiley-lAss, New York. pp. (,q -IIIIL Roberts, J.l).. D.C. Thomas and T.A. Kunkel (19(I11 I(xonucleolytic proofreading of leading and lagging strand I)NA replication errors, Proc. Natl. Acad. Sci. (I.;.S.A.L NS. 3465-3469. "]'hom~.s, 1).('., J.l). Roberts and T.A. Kunkel (1¢)~/1) Itcteroduplex repair in extracts of human t lel,a cells, J. Biol. ('hem., 266, 3744-3751. Tsurimolo. T.. T. Melendy and B. Stillman (19u)()) Sequential initiation of lagging and leading strand ,',ynthesis by two different polymerase complexes at the SV40 DNA replication origin, Nature (LondonL 346, 534-539. Wong, S.W., I,.R. Paborsky. P.A. Fisher, T.S.-F. Wang and 1). Korn (19861 Structural and cnzymological characterization of immunoaffinity-purificd DNA polymcrase ¢~'I)NA primast complex from KI3 cells, J. Biol. ('hem, 261, 7c,~58 7968.
175
© 1991 Elsevier Science Publishers B.V. All rights reserved IX127-5107/91/$03.50 ADONIS 002751079100176F
MUT 02510
The fidelity of DNA synthesis by the catalytic subunit of yeast DNA polymerase a alone and with accessory proteins Thomas A. Kunkel, John D. Roberts and Akio Sugino Laboratory of Molecular Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 (U.S.A.) (Accepted 5 April 1991)
Keywords: Fidelity; DNA polymerase a; DNA replication
Summary The fidelity of DNA synthesis catalyzed by the 180-kDa catalytic subunit (p180) of DNA polymerase a from Saccharomyces cerecisiae has been determined. Despite the presence of a 3 ' ~ 5' exonuclease activity (Brooke et al., 1991, J. Biol. Chem., 266, 3005-3015), its accuracy is similar to several exonuclease-deficient DNA polymerases and much lower than other DNA polymerases that have associated exonucleolytic proofreading activity. Average error rates are 1/9900 and 1/12000, respectively, for single base-substitution and minus-one nucleotide frameshift errors; the polymerase generates deletions as well. Similar error rates are observed with reactions containing the 180-kDa subunit plus an 86-kDa subunit (p86), or with these two polypeptides plus two additional subunits (p58 and p49) comprising the DNA primase activity required for DNA replication. Finally, addition of yeast replication factor-A (RF-A), a protein preparation that stimulates DNA synthesis and has single-stranded DNA-binding activity, yields a polymerization reaction with 7 polypeptides required for replication, yet fidelity remains low relative to error rates for semiconservative replication. The data suggest that neither exonucleolytic proofreading activity, the /3 subunit, the DNA primase subunits nor RF-A contributes substantially to base substitution or frameshifl error discrimination by the DNA polymerase a catalytic subunit.
DNA replication in extracts of human (Roberts and Kunkel, 1988) or monkey (Hauser et al., 1988) cells is highly accurate. In human cell extracts, this accuracy results both from high nucleotide selectivity and from exonucleolytic proofreading on both the leading and the lagging strand (Roberts et al., 1991). One approach for under-
Correspondence: Dr. Thomas A. Kunkel, Laboratory of Molecular Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709 (U.S.A.), Phone: (919)-541-2644; Fax: (919)-541-7593.
standing the molecular details of mutational processes during replication is to define the proteins responsible for these two discrimination steps. The observations that lagging strand replication is accurate and that exonucleolytic proofreading occurs on the lagging strand are striking when considered in the context of the current model for the eukaryotic replication fork. This model, which is supported by a large body of evidence in vitro (e.g., Kelly, 1988; Tsurimoto et al., 1990; Hurwitz et al., 1990), posits that DNA polymerase a is responsible for lagging strand synthesis. However, mammalian DNA poly-
176 mcrase c~ preparations have historically (for rcview, see Fry. and I_,mb, 1986) bccn found Io be devoid of 3 ' ~ 5' exonuclease activity, although exceptions have been noted (see below}. Furthermore. based on our experience with preparations from a variety of sources (Kunkel, 1985: Roberts and Kunkel, 1988: Kunkel et al.. 1989), the fidelity of D N A synthesis by D N A polymcrasc (~ is consistently much lower than that of lagging strand replication (Roberts et al., 1991 ). Given these observations, it is reasonable to ask whether more accurate forms of D N A polymerase ~e can bc found, tligher accuracy might result if the polymerasc has an associated 3' --* 5' exonucleasc a n d / o r additional associated proteins that could enhance nucleotide selectivity. An ideal place to begin thc scarch for such fidelity factors is in Saccharomvces cererisiue. This organism not only provides a powerful genetic system, but is similar to mammalian systcms regarding its replication enzymology (e.g., Burgers. 1989: Brill and Stillman, 1989). This similarity includes the involvement of D N A polymerasc ~ (formerly designated D N A poIymerase I) as an esscntial rcplicative polymerasc. The catalytic subunit of yeast I)NA polymerase Ce, like its mammalian counterpart, is a 180-kDa polypeptide (Plevani et al., 1985). When overproduccd in yeast from the cloned gene and then extensively purified (Brooke et al., 1991), this plS0 protein not only has polymerization activity but also 3'--~ 5' exonucleolytic activity, effectively digesting a hmg, synthetic, singlestranded po[ynucleotide substrate. This exonuc[ease is thus a candidate for a proofreading activity. Similar to its mammalian equivalent, yeast I)NA polymerase ~r can be purified by immunoaffinity chromatography as a four-subunit complex (for revicw, see Burgers, 1989). In addition to the catalytic p180, this complex contains an 86-kDa polypeptide (p86) that has no apparent enzymatic activity but stabilizes the polymerase activity against thermal inactivation (Brooke ct al., 1991). The g e n t coding for p86 has also been cloned, allowing overproduction of the protein in yeast and subsequent extensivc purification (Brooke ct al., 1991). The D N A polymerasc a complex also contains two polypeptides (58- and 49-kDa, designated p58 and p49, respec-
lively) that together comprise a DNA primasc activity for initiating lagging strand ()kazaki fragnlents.
The similarity of the yeast and mammalian replication systems extends to other proteins required for scmiconservative I)NA replication. Among these is a yeast single-stranded I)NAbinding protein that can participate in origin-dependent initiation (Brill and Stillman, 1989). The preparation actually consists of three polypcptides (69, 3¢~ and 13 kDa), collectively referred to as replication factor-A (RF-A). In addition to its role in intiation, RF-A is thought to participate in chain elongation reactions on I~oth the leading and lagging strands. As prokarvotic singlestranded I)NA binding proteins have been shown to intluence the fidelity of DNA synthesis with purified DNA polymerases (Kunkcl et al.. 1979L it was of interest to examine the influence of yeast RF-A on the fidelity of yeast I)NA polymeras¢ ¢~. In the present study, wc have used assays that monitor base-substitution and frameshift error rates during a single cycle of DNA synthesis in vitro (Kunkel ct al., 1989), to examine the fidelity of piN0 alone or in the presence of pN6, p86 plus p58 and p49, and p86, p58, p49 and R[:-A. The data suggest that, despite the presence of a 3' --, 5' exonuclease activity and six additional proteins, DNA synthesis catalyzed by the catalytic subunit of yeast DNA polymerase ~ is not highly accurate, being much less accurate than lagging strand I)NA replication in cxtracts of human cells. Materials and methods
Materials The bacterial strains, bacteriophage M l 3 m p 2 and its derivatives and other reagents required for the fidelity assays have been described (Kunkel et al,, 1989 and refs. therein). The p180 and p86 proteins were kindly provided by G. Brooke (University of Illinois at Champaign-Urbana). These were overproduced, purified and characterized as described (Brooke el al., 1991). DNA polymerase-¢~-DNA primase was immunoaffinitypurified as described (Kunkcl et al., 1989). The purification and characterization of yeast RF-A will be described elsewhere (A. Sugino, in prepa-
177
ration). Its properties and polypeptide compostion are similar to those described by Brill and Stillman (1989).
Methods Ml3mp2 mutagenesis assays. Three assays for measuring the fidelity of a single round of gapfilling DNA synthesis were employed. In all three cases, a gapped, double-stranded Ml3mp2 DNA substrate is constructed (Kunkel and Soni, 1988) such that the single-stranded gap contains the mutational target. (i) Forward-mutation assay. The forward-mutation assay (Kunkel, 1985) uses thc wild-type lacZa sequence. Correct polymerization to fill the 390-base gap produces DNA that, when used to transfect an E. coli host strain, will produce dark blue M13 plaques. Errors during synthesis are scored as lighter blue or colorless plaques. Since the assay measures loss of a gene function (a-complementation of /3-galactosidasc activity) that is not essential for phage production, a wide variety of mutations at many different sites can be recovered and scorcd. This includes 227 differcnt single base substitution errors at 116 different template positions, single-base frameshifts at 150 different template positions involving both nonreiterated and reiterated bases, deletions and complex errors. (ii) Opal codon reversion assay. It is possible to focus exclusivcly on single-base substitution errors using the opal codon reversion assay (Kunkel and Soni, 1988). In this assay thc 361-base gap contains a single base change [G ~ A in the viral (plus) template strand at position 89 of the lacZa coding sequence]. This change creates an opal (TGA) codon, resulting in a colorless-plaque phenotype. Base-substitution errors resulting from gap-filling synthesis are detected as blue plaques and the reversion frequency (the proportion of blue to total plaques) reflects the base substitution error rate. 8 of 9 possible basc-substitution errors at the TGA opal codon yield a dctectable bluc plaque phenotype. (iii) Assay for minus-one base frameshifts. This assay (Bebenek and Kunkel, 1990) also employs an M13mp2 DNA substrate containing a 361-base gap and also scores blue revertants of a colorless
plaque phenotype. In this instance one begins with a mutant that is colorless due to the addition of one extra base in the lacZa coding sequence, a T into the run of four consecutive Ts at positions 70-73 (using our previous convention (22) this is referred to as the + T70 mutation). Blue plaques result from errors that restore the reading frame.
p180-p86 complex formation. 13 ng each of p180 and p86 were incubated for 2 h at 0 ° C in 10 mM Hepes (pH 7.4), 2.5 mM dithiothreitol. This has been shown to yield complex formation as determined by co-immunoprecipitation of the two polypeptides (Brooke et al., 1991). DNA polymerase reactions. Reactions (50 gl) contained 20 mM Hepes (pH 7.4), 2 mM dithiothreitol, 6 mM MgCI 2, 6.5% (for p180 or polymerase-primase) or 13% (p180-p86) glycerol, 10 /zg BSA, 150 ng gapped M13mp2 DNA (either wild-type or mutant, depending on the assay), either 13 ng of p180, 26 ng of p180-p86 complex or 5 units of polymerase-primase complex, and, unless noted otherwise, 500 /zM each of dATP, dTI'P, d G T P and dCTP. After incubation at 30 ° C for 2 h, reactions were terminated by addition of E D T A to a final concentration of 1(1 mM. 15-p.l aliquots of each reaction were analyzed by agarose gel electrophoresis as described (Kunkel, 1985). For all conditions, the gap was filled to the extent that the DNA migrated coincident with a fully double-stranded, nicked circular (RFII) DNA standard (not shown). Other procedures. Preparation of M13mp2 DNA substrates, transfection of competent ceils, plating and scoring of o~-complcmentation mutants and DNA sequence analysis were performed as described (Kunkel et al., 1989 and refs. therein). Results
We recently demonstrated (Kunkel et al., 1989) that yeast DNA polymerase o~, whether purified by conventional chromatography or by immunoaffinity chromatography as a polymeraseprimase complex, did not excise a 3'-OH-terminally mispaired nucleotide from either of two
178
different DNA substrates, even after extended incubation with a substantial excess of polymerase. This lack of exonucIeolytic activity was confirmed by Brooke et al. (1991), who demonstrated that plS0, with or without the additional polypeptides of the polymerase-primase complex, failed to degrade the radiolabeled oligo (dC)l.~ mispaired portion of an oligo(dA)ts(dC)~ -poly (dT) substrate, under conditions in which several DNA polymerases containing proofreading activity readily digested this substrate. However, Brooke et al. (1991) also found that all three forms of yeast DNA polymerase o~ (plS(I, p180-p86 and the polymcrase-primase complex) released dCMP from the 3'-OH terminus of a single-stranded s u b s t r a t c , poly(dT)~,0~j-32 PdCMP0. 4. The exonuclease and polymerasc activities co-immunoprecipitated using antibodies directed against either p18(1 or p86, suggesting an association between the two activities. The detection of this polymcrase-associated 3' ~ 5' cxonuclease activity prompted us to examinc the fidelity of all three forms of yeast DNA polymerase ot under conditions where proofreading activity might be expected to yield high fidelity DNA synthesis. For this purpose we used the opal codon reversion assay (sec Methods) for base-substitution errors that arc known to be proofread effectively by a variety of DNA polymerases having associated exonuclease activity. Gap-filling synthesis reactions contained relatively low concentrations of
dNTPs, in order to detect any proofreading activity that might be present. As expected, DNA synthesis by enzymes having associated proofreading activity was highly accuratc. Low reversion frequencies were observed (Table 11 for the large fragment of E. coli DNA polymerase i (Bebenek and Kunkel, 1990), bovine DNA polymerase • (formerly designated DNA polymerase 611, Kunkel et al., 1987), chick-embryo DNA polymcrase y (Kunkel and Soni, 1988), porcineliver DNA polymerase y (Kunkel and Mosbaugh, 1989) and T7 DNA polymerase (data not shown). In contrast, the catalytic subunit of yeast DNA polymcrase o~ generated revertants at a frequency that was - 100-fold above the background frcquency of uncopied DNA. Similar fidelity was observed when p86 was present or when the intact, immunoaffinity-purified DNA polymerasc ~e-DNA primasc complex (Table 1 and Kunkel ct al., 19891 was employed. The fidelity of these forms of yeast pol o~ was comparable to that of the cxonuclease-deficicnt form of the large fragment of E. coli DNA polymerase 1 (Bcbcnek and Kunkel, 199(I, and Tablc 1). The error ratc per dctectable nucleotidc polymerized, ~ 1/100(10, is comparable to the estimated base selectivity of a variety of exonuclcasc-deficient DNA polymcrascs (for review, scc Echols and Goodman, 1991). The data suggest that the exonuclease activity detected that releases dCMP from singlestranded poly(dT)~,~-~-'P-dCMP~,, does not contribute significantly to the base-substitution fi-
TABLE 1 Base-substitution fidelity of yeast plS0, p180-p86 and polymerase-primasc compared to that of DNA Ix)lymerases known t¢) proofread errors DNA polymerase
3' --, 5'
Plaques scored
used
exonuclease activity
Total
Mutant
frequency ( × 10 - ~')
F. coli Pol I, large fragment bovine DNA polymerase ~ chick DNA polymerase y porcine DNA polymerase y yeast p180 yeast p 180 + p86 yeast polymerase-primase E. coli Pol l, large fragment
+ + + +
3 700(l~10 2 3000(,Ill 600000 6 100000 1100(X) 14(}000 730000 460{XXI
69 12 5 32 21 15 78 56
18 5.2 8.3 5.2 190 1 Ill 110 120
-
The background mutant frequency of uncopied DNA was 2 × 10 -~'.
Reversion
179 TABLE 2 Fidelity of yeast p180, p l S 0 - p 8 6 and polymerase-primase in the forward mutation assay D N A polymerase
Plaques scored
used
Total
Mutant frequency ( × 111-4)
Mutant
Expt. I p180 p 180 + p86 Polymerase-primase (Kunkel et al., 1989)
2591 4 ill 3
55 92
6 328 6 284 6 409
279 296 270
210 230 140 +_20
Expt. 2 p 180 p 180 + p86 Polymerase-primase
440 47(1 420
The background mutant frequency was 6 . 7 x 10 4
delity of gap-filling synthesis by yeast DNA polymerase a under the reaction conditions employed in this study. The opal codon reversion assay monitors single base-substitution errors at three template nucleotide positions. Since DNA polymerase error rates have been found to depend on the local sequence around the error as well as on the composition and symmetry of the error (for review, see Kunkel and Bebenek, 1988), we next examined the fidelity of the three yeast pol a forms in the M13mp2 forward mutation assay to see if differences could be found. This assay scores a wide variety of mutations at many different sites (see Methods). As for the reversion assay, all three forms of polymerase had comparable accuracy, in two independent experiments (Table 2). Again, the forward mutant frequency values are comparable to those generated by several DNA polymerases that lack proofreading activity (Kunkel, 1985; Roberts and Kunkel, 1988; Kunkel et al., 1989). DNA sequence analysis of 134 mutants generated by p180 and of 20 mutants generated in reactions containing p180 plus p86 (Table 3) demonstrate that both forms of polymerase generate single base-substitutions, single-base frameshifts and deletions of larger numbers of nucleotides, that the two forms of polymerase have comparable accuracy, and that their error rates and error specificities are remarkably similar to those of the DNA polymerase a-primase complex described earlier (Kunkel et al., 1989).
In considering which accessory proteins might contribute to the high fidelity of semiconservative replication, an attractive candidate is yeast RF-A. This three-subunit yeast protein preparation stimulates synthesis by yeast DNA polymerase a (not shown) and the largest subunit (69 kDa) has single-stranded DNA-binding activity (Brill and Stillman, 1989). In addition to its suggested role in initiation at a replication origin (Brill and Stillman, 1989), yRF-A has been suggested to participate in the elongation phase of the reaction. We therefore asked if addition of yRF-A to reactions catalyzed by the presumed replicative form of yeast pol a, the DNA polymerase otprimase complex, altered fidelity. The addition of yRF-A had no apparent effect
TABLE 3 Types of errors as defined by D N A sequence analysis Type of
p180
mutation
NumM.F. NumM.F. b e r o f ( x l 0 -4) b e r o f (xl0 mutant mutants
B a ~ substititions 44 O n e base frameshifts 48 Deletions > 1 base 31 Other a 11 Total 134
plS0-p86
69 75 49 17 210
7 7 3 3 20
4)
80 80 35 35 230
a This class includes mutants containing more than a single nucleotide difference from wild-type or no changes within the lacZa sequence present within the gapped D N A substrate.
180 TABI ,E 4 Fidelity of yeast [)NA polymcrasc (r-primase with and without RF-A Components
Plaques scored
used
Total
Mutant
Mtlt[mt frequency { × I0
a)
E3cpt, 1. Forward Mutation Assa.v Polymerase- primasc Polymcrase. primase plus yRF-A
6 572 5 167
t)l 79
1411 1511
9901Xl
1112 1211
lilO I(H)
l'[xpt. 2. Mimt.s-one Nlwleotlde Ret'erstotl .,'|.s.';ay Polymerase -primase Polymerase -primase plus yRF-A
12()01X)
The background mutant frequency for the forward assay was 6.7 × 10 4 and for the reversion assay was 2,4 × l(I 5 For reactions containing yRF-A, 2 # g was added.
in the forward mutation assay (Table 4, Expt. 1). Among the mutants recovered, the ratio of light blue to colorless mutants was similar, providing suggestive evidence that yRF-A was not strongly influencing error specificity. Rather than sequencing another large collection of mutants to examine this further, wc determined the effect of addition of yRF-A on the frameshift fidelity of the polymerase-primase. Using a reversion assay (Bebenek et al., 1990) that scores minus-onenuclcotide errors at a template T V I T I ' sequencc (as dark blue revertants) or at 38 non-reiterated template nucleotide positions (as lighter blue revertants), the addition of yRF-A was found to have no effect on thc frameshift fidelity of thc polymerase oe-primase complex (Table 4, Expt. 2). Discussion The major conclusion from the present study is that the average fidelity of DNA synthesis catalyzed by replicative yeast DNA polymerase a is low relative to the estimated accuracy of semiconservative replication, despite the prescncc of 6 additional replication proteins and the reported presence of an associated 3'--*5' exonuclease activity. These data suggest that the primary determinants of selectivity against base-substitution and frameshift errors arc present in the p180 catalytic subunit itself, and that, at least in the in vitro system used here, the exonuclease is not contributing significantly to the fidelity of DNA polymcrasc oe. The latter conclusion is consistent
with the known DNA substratc requirement fl)r the 3 ' - - , 5 ' exonucleasc activity in the plSI) preparation. Thc exonuclcase only degraded a long singlc-strandcd polynucleotide but did not excise a terminal mismatch from a primer-template (Brooke et al., 1991), as has bccn observcd with 3' ---, 5' exonucleascs that proofread errors. The lack of an effect on fidelity with p86 and the primase subunits is consistent with numerous reports that these subunits of DNA polymerase ce-primasc complexes have littlc effect on other properties of the polymerization reaction (Wong et al., 1986; Cottcrill et al., 1987; Brooke el al.. 1991). Givcn the ability of prokaryotic singlestranded binding proteins to altcr the base-substitution error rates of DNA polymerases in vitro (Kunkel et al., 1979), the lack of an cffcct by yRF-A in either the forward mutation assay or the frameshift reversion assay was somewhat unexpected. At least for minus-one nucleotide frameshift errors at the T T I T F run (Table 4, Expt. 2), this lack of an effcct is not unreasonable. Such errors probably result from templateprimer misalignments whcrein the extra template base would be present in an otherwise doublcstranded primer stem (c.g., see Fig. 3 in Kunkel. 1990) and therefore unavailable to the singlestranded binding protein. The absence of an yRF-A effect on polymerasc fidelity may reflect the possibility that it could serve a diffcrent role in replication than has been attributed to prokaryotic SSBs. For example, its role may be limited to initiation of replication at the origin,
181
while another SSB participates in the chain elongation phase of replication. It is of course possible that the absence of fidelity-enhancing effects may reflect inappropriate reaction conditions or less than ideal protein-protein interactions. Thus, these proteins could still be important fidelity components in the context of a more complete replication apparatus. It is also possible that these proteins could improve fidelity for certain types of errors (by position or composition) that have not yet been exhaustively examined in the present study. For example, it is not unreasonable to expect that a protein like yRF-A that can bind to singlestranded DNA might reduce the frequency of deletion errors between directly repeated DNA sequences that are known to result from polymerization by DNA polymerase a (Table 3 and Kunkel et al., 1989). There is in fact precedent for this; when a 38-kDa SSB from yeast (one now known to probably be unimportant for DNA replication) was included in reactions catalyzed by yeast DNA polymrase a, fidelity was increased for deletions between direct repeats (Roberts et al., 19901. This possibility could be tested with yRF-A by sequence analysis of a large collection of mutants generated in the forward mutation assay. The fidelity studies reported here may have relevance beyond understanding error discrimination during semiconservative DNA replication. Substantial evidence supports a role for DNA polymerase a in repair of mismatched bases (Holmes et al., 1990; Thomas et al., 1991) and DNA damage (for review, see Fry and Loeb, 1986). The subunit composition of DNA polymerase a serving these roles may well be different. For example, the primase subunits may not be associated with the polymerase catalytic subunit. In the absence of additional information, the fidelity of the 180-kDa catalytic subunit alone reported here could be considered a baseline for comparison to future studies of the fidelity of DNA polymerase a-dependent repair synthesis.
Acknowledgements We would like to thank Glenn Brooke for generously providing p180 and p86 proteins and
Kenneth Tindall and James Clark for critical evaluation of the manuscript.
References Bebenek, K., C.M. Joyce, M.P. Fitzgerald and T.A. Kunkel (19901 The fidelity of DNA synthesis catalyzed by derivatives of Escherichia coli DNA polymerase I, J. Biol. Chem., 265, 13878-13887. Brill. S.J., and B. Stillman 119891 Yeast replication factor-A functions in the unwinding of the SV40 origin of DNA replication, Nature (lamdon), 342, 92-95. Brooke, R.G., R. Singhal, D.C. Hinkle and L.B. Dumas 119911 Purification and characterization of the 1811- and 86-kilodalton subunits of the Saccharomyces cereHsiae DNA primase-DNA polymerase protein complex, The 180 kilodalton subunit has both DNA polymerase and 3' --* 5' exonuclease activities, J. Biol. Chem., 266, 3005-31115. Burgers, P.J. 11989) Eukaryotic DNA polymerases a and h: Conserved properties and interactions, from yeast to mammalian cells, Progress in Nucleic Research and Molecular Biology, 37, 235-280. Cotterill, S.M., M.E. Reyland, L.A. Loeb and I.R. Lehman (19871 A cryptic proofreading 3' --, 5' exonuclease associated with the polymerase subunit of the DNA I~lymeraseprimase from Drosophila melanogaster, Proc. Natl. Acad. Sci. (U.S.A.). 84, 5635-5639. Echols, H., and M.F. Goodman (1991) Fidelity mechanisms in DNA replication, Annu. Rev. Biochem., in press. Fry, M., and L.A. Loeb (1986) in: Animal cell DNA polymerases. CRC Press, Inc., Boca Raton, FL. Hauser, J., A.S. Levine and K. Dixon (1988) Fidelity of DNA synthesis in a mammalian in vitro replication system, Mol. Cell. Biol., 8. 3267-3271. Holmes Jr., J., S. Clark and P. Modrich (1990) Strand-specific mismatch correction in nuclear extracts of human and Drosophila melanogaster cell lines, Proc. Natl. Acad. Sci. (U.S.A.), 87, 5837-5841. tlurwitz, J., F.B. Dean, A.D. Kwong and S.-ll. Lee (1990) The in vitro replication of DNA containing the SV4(1 origin, J. Biol. Chem., 265, 18043-18046. Kelly, T.J. (1988) SV40 DNA replication, J. Biol. Chem., 263, 17889-17892. Kunkel, T.A. (19851 The mutational specificity of DNA I:x~lymerase-/3 during in vitro DNA synthesis, Production of frameshift, base substitution and deletion mutations. J. Biol. Chem., 260, 5787-5796. Kunkel. T.A., and K. Bebenek (1988) Recent studies of the fidelity of DNA synthesis, Biochim. Biophys. Acta. 951, 1-15. Kunkel, T.A., and D.W. Mosbaugh (1989) Exonucleolytic proofreading by a mammalian DNA polymerase gamma, Biochemistry, 28, 988-995. Kunkel, T.A., and A. Soni (19881 Exonucleolytic proofreading enhances the fidelity of DNA synthesis by chick embryo DNA polymerase-y, J. Biol. Chem., 263, 4450-4459.
182 Kunkel, T.A., R.R. Meyer and L.A. Lx~eb(1979) Single-strand binding protein enhances fidelity of DNA synthesis in vitro, Proc. Natl. Acad. Sci. (U.S.A.). 76, 6331-6335. Kunkel, T.A.. R.D. Sabatino and R.A. Bambara (19871 Exonucleolytic proofreading by calf thymus I)NA polymerase 6, Proc. Natl. Acad. Sci. (U.S.A.), 84. 4865 4869. Kunkel, T.A., R.K. Hamatakc, J. Motto-Fox, M.P. Fitzgerald and A. Sugino (19891 The fidelity of DNA polymerasc I and the DNA polymerase I-DNA primasc complex from Saccharomyces cererisiae, Mol. Ceil. Biol.. 9, 4447-4458. Plcvani, P., M. Foiani, P. Valsasnini, G. Badaracco, IL Cheriathundam, and I,.M.S. Chang (19851 Polypcptidc structure of DNA primase from a yeast DNA polymerase primase complex, J. Biol. ('hem.. 26(I, 71112.-71117. Roberts, J.D., and T.A. Kunkcl (1988) Fidelity of a human cell DNA replication complex, Prtvc. Natl. Acad. Sci. (U.S.A.), 85, 7064-7068. Roberts, J.D., R.K. Hamatakc, M.S. Fitzgerald, A. Sugino, and T.A. Kunkel (19901 Effect of accessory' proteins on the fidelity of DNA synthesis by eukaryotic rcplicative poly-
merases, in M.L. Mcndelsohn and R.J. Aibertini (l{ds.). Progress in Clinical and Biological Researcll. Vol. 34{)A, Mutation and the Environment, Part A: Basic Mechanisms, Wiley-lAss, New York. pp. (,q -IIIIL Roberts, J.l).. D.C. Thomas and T.A. Kunkel (19(I11 I(xonucleolytic proofreading of leading and lagging strand I)NA replication errors, Proc. Natl. Acad. Sci. (I.;.S.A.L NS. 3465-3469. "]'hom~.s, 1).('., J.l). Roberts and T.A. Kunkel (1¢)~/1) Itcteroduplex repair in extracts of human t lel,a cells, J. Biol. ('hem., 266, 3744-3751. Tsurimolo. T.. T. Melendy and B. Stillman (19u)()) Sequential initiation of lagging and leading strand ,',ynthesis by two different polymerase complexes at the SV40 DNA replication origin, Nature (LondonL 346, 534-539. Wong, S.W., I,.R. Paborsky. P.A. Fisher, T.S.-F. Wang and 1). Korn (19861 Structural and cnzymological characterization of immunoaffinity-purificd DNA polymcrase ¢~'I)NA primast complex from KI3 cells, J. Biol. ('hem, 261, 7c,~58 7968.
