Studies on the DNA Elongation Inhibitor and Its Proliferating Cell

Proc. Nati. Acad. Sci. USA Vol. 86, pp. 4877-4881, July 1989 Biochemistry Studies on the DNA elongation inhibitor and its proliferating cell nuclear antigen-dependent control in simian virus 40 DNA replication in vitro (DNA polymerases/termini of DNA/regulation of DNA synthesis) SUK-HEE LEE, ANN D. KWONG, YUKIO ISHIMI, AND JERARD HURWITZ Graduate Program in Molecular Biology, Memorial Sloan-Kettering Cancer Center, New York, NY 10021 Contributed by Jerard Hurwitz, March 28, 1989

ABSTRACT A 120-kDa protein that blocks DNA termini circular 4X174 DNA was provided by K. Marians (Memorial has been purified from extracts of HeLa cells. This protein Sloan-Kettering Cancer Center) and a OX174 primer (30- inhibits the action of a number of enzymes that catalyze mer, nt 5127-5156) was synthesized by using an Applied reactions involving the 5' and 3' ends of DNA (DNA ligase, 3' Biosystems 380A DNA synthesizer. Enzymes and fractions and 5' exonucleases, and DNA polymerase a). The 120-kDa isolated from HeLa cell extracts included immunopurified protein blocks the synthesis of long DNA chains that are DNA polymerase a-primase (1.6 x 103 and 1.9 X 103 normally formed during simian virus 40 DNA replication, units/mg of protein, respectively), HeLa SSB (632 units/mg causing the accumulation of small DNA fragments. The effects of protein), 0.3 M NaCl eluate from phosphocellulose (PC) of this protein can be reversed by the addition of proliferating (DNA polymerase a-primase; 18 and 50 units/mg of protein, cell nuclear antigen and other protein fractions (activators). respectively), topo 1 (5.0 x 106 units/mg of protein), 5' to 3' exonuclease (1.4 x 105 units/mg ofprotein), DNA ligase (0.8 The replication of simian virus 40 (SV40) DNA has been unit/mg of protein), and PCNA (5.2 x 105 units/mg of extensively reviewed (1, 2), Since the establishment of an in protein). The above were prepared as described (11, 14). vitro SV40 replication system (1-5), a number of well- Escherichia coli exonuclease III was purchased from Boe- characterized proteins have been shown to play a direct role hringer Mannheim Biochemicals. in SV40 DNA replication. In addition to the SV40-encoded Assays ofEnzymatic Reactions and the Elongation Inhibitor. large tumor antigen (SV40 T antigen) (3), proteins that have The 5' to 3' exonuclease and DNA ligase were assayed as been purified from human cells include DNA polymerase a described (11). The substrate used for the assay of exonu- and DNA primase (6), topoisomerases (topo) I and II (7), and clease III was prepared as follows: 160 pmol of 4X174 a single-stranded (ss) DNA-binding protein (SSB) (8-10). oligomer (30-mer, nt 5127-5156) was annealed to the 4X174 RNase H, a 5' to 3' exonuclease, and DNA ligase have been ss circular DNA (1.8 pmol) and 3'-end labeled by using shown to be required for the formation of closed circular [a-32P]dATP and the Klenow fragment of E. coli DNA duplex DNA (11). polymerase I (15). The exonuclease III elongation inhibitor Proliferating cell nuclear antigen (PCNA), a putative cell- assay was carried out as follows: reaction mixtures (20 1l) cycle-regulated protein, was shown to be essential in an SV40 containing 7.7 fmol of 3' end-labeled singly primed 4X174 in vitro replication system containing relatively crude frac- DNA (2000 cpm/fmol of molecules), 0.8 gg of HeLa SSB, 50 tions (12). Furthermore, PCNA was required for leading- mM Tris HCI (pH 7.5), 0.67 mM MgC12, bovine serum strand DNA synthesis, since only small DNA fragments albumin (25 gg/ml), and 1 mM dithiothreitol were incubated arising from the lagging strand accumulated in its absence with fractions containing the elongation inhibitor. After 15 (13). In contrast, PCNA had no effect in a reconstituted min at 250C, 0.01 unit of exonuclease III was added, and the system containing T antigen, HeLa SSB, topo I, and the reactions were incubated for 30 min at 370C. Reactions were DNA polymerase a-primase complex as purified compo- stopped by adding 2 jul of 50 mM EDTA and were analyzed nents (14). The different requirement for PCNA was traced to by 7 M urea/8% PAGE or by quantitation of the amount of an elongation inhibitor protein, which was present in the alcohol-soluble [32P]dAMP released by exonuclease III. crude fractions but not in the purified reconstituted system. The DNA polymerase a elongation assay used the same The elongation inhibitor prevented the synthesis oflong DNA singly primed 4X174 DNA substrate as the exonuclease III chains, which resulted in the accumulation of short DNA assay. Reaction mixtures (30 IlI) containing 12.5 fmol of the fragments [average length, 300 nucleotides (nt)]. Its effect 4X174 DNA-primed template, 20 mM Tris HCl (pH 7.8), 5 could be reversed by PCNA and protein activators (14). mM MgCl2, 0.8 mM dithiothreitol, bovine serum albumin (75 In this report, we present the characterization of the ,g/ml), 0.6 ,ug of HeLa SSB, 150 AM dCTP, 150 ,M dATP, elongation inhibitor protein. In addition, we have examined 150 ,M dGTP, 50 ,uM [3H]dTTP (200 cpm/pmol), and 0.5 mM a number of enzymatic reactions that are affected by this ATP were incubated with fractions containing the elongation inhibitor and have identified protein activator fractions that inhibitor. After 5 min at room temperature, 0.2 unit of are essential for the reversal of the effects of this inhibitor. immunopurified DNA polymerase a was added, and reactions were incubated for 30 min at 37°C. The amount of MATERIALS AND METHODS acid-insoluble radioactivity was then determined. The inhib- Preparations of Proteins and Nucleic Acids. Cytosolic ex- Assay and Isolation of the Elongation Inhibitor. tracts of HeLa cell, SV40 ori+ DNA (pSV01AEP), and SV40 itor was assayed by its ability to decrease deoxynucleotide T antigen were prepared as previously described (3). ss Abbreviations: SV40, simian virus 40; SV40 T antigen, SV40- encoded large tumor antigen; SSB, single-stranded DNA-binding The publication costs ofthis article were defrayed in part by page charge protein; ds, double-stranded; ss, single-stranded; topo, topo- payment. This article must therefore be hereby marked "advertisement" isomerase(s); PCNA, proliferating cell nuclear antigen; nt, nucleo- in accordance with 18 U.S.C. §1734 solely to indicate this fact. tide; PC, phosphocellulose.

4877 Downloaded by guest on September 29, 2021 4878 Biochemistry: Lee et al. Proc. Natl. Acad. Sci. USA 86 (1989) incorporation during the replication of SV40 DNA in vitro. 10% (vol/vol) glycerol/0.1 mM phenylmethylsulfonyl fluo- One unit of inhibitor activity was defined as the amount ride/0.01% Nonidet P-40/antipain at 0.2 pug/ml) plus 0.2 M required for a 50% reduction in nucleotide incorporation. In NaCl. The column was washed successively with 180 ml of addition to nucleotides and the ATP-regenerating system buffer A containing 0.2, 0.3, 0.4, 0.7, and 1 M NaCl. The (14), reaction mixtures (40 ttl) contained 0.6 1Lg of SV40 T inhibitor eluted with the 0.7 M NaCl wash, which was then antigen, 0.8 Ag of HeLa SSB, 10 units of topo I, 0.23 Ag of dialyzed against 4 liters of buffer A containing 20% (wt/vol) pSV01AEP (RFI), 6 ,ug of 0.3 M NaCl PC fraction, which sucrose and 25 mM NaCl (45 ml, 107 mg of protein, 4.5 x 104 contained DNA polymerase a and primase (14), and the units; specific activity, 420 units/mg ofprotein). This fraction inhibitor fraction. After 60 min at 370C, aliquots were re- was loaded onto a ssDNA-cellulose column (8 mg of protein moved for gel electrophoresis of products and the determi- per ml of bed volume; 1.8 x 5.3 cm) equilibrated with buffer nation of acid-insoluble material as described (14). A plus 0.2 M NaCl. The column was eluted successively with Cytoplasmic extracts of HeLa cells (100 ml, 30 mg of buffer A (45 ml each) containing 0.2, 0.4, 0.8, and 2.0 M NaCl. protein per ml) were adjusted to 0.2 M NaCl and loaded on The elongation inhibitor eluted at 0.8 M NaCl. This fraction a PC column (2.5 x 17 cm; 85 ml) equilibrated with buffer A was dialyzed against buffer A containing 20%o sucrose and 25 (25 mM Tris-HCl, pH 7.5/1 mM dithiothreitol/1 mM EDTA/ A a 12 16 202428 32 3640444852 56 60 64 A 200- PCNA (ng) 10 20 40 80 160. 116- 0.4 M ds DNA ()Lg) - 0150.3 0.6 1.2 1.2 1.2 1.2 1.2 1.2 97,- 66-

240 2480 - 45 - 32-

B dNMP incorp (pmol) 70 19 63 64 61 52 43 23 28 33 38 50 56 60 69 66 Fraction Ca b 12 16 20 2428 32 36 40444852 566064

2480 - 310 -

310 - PCNA added(QLg) + Inhibitor(1.2pLg) 0.02 0.04 0.06 0.08 0.10 0.16

C c 12 16 20 24 2832 36404448 52 5660 64 q E 32mer- o 0L -

dAMP - _O

C FIG. 2. Chromatography ofthe elongation inhibitor on a dsDNA- cellulose column. The dsDNA inhibitor fraction (0.65 mg of protein) z was applied to a dsDNA-cellulose column (0.8 x 1.0 cm) equilibrated with buffer A plus 50 mM NaCI. The column was eluted with 10 ml of a 0.05-1 M NaCl linear gradient in buffer A. The NaCl concentrations of fractions 20, 30, 40, and 50 were 0.15, 0.3, 0.48, and 0.65 M, respectively. (A) SDS/PAGE analysis of the gradient fractions. The numbers at the top of the lanes indicate the fractions of the 0.2 0.4 0.6 0.8 1.0 1.2 gradient analyzed. The positions of size markers (in kDa) are 0.4 M ds DNA Inhibitor fraction(Qg) indicated at the left. Lane a denotes the starting dsDNA-cellulose fraction (0.6 pg). (B) Effect of fractions on SV40 DNA replication. FIG. 1. Effects of the elongation inhibitor and PCNA on the Reaction mixtures containing 4 Al of each fraction were as described replication of SV40 ori+ DNA. The influence of increasing amounts in Materials and Methods, and the products were analyzed as of the 0.4 M NaCl dsDNA-cellulose fraction of the elongation described in Fig. 1A. Lane a contained no elongation inhibitor inhibitor and purified PCNA were examined in the SV40 replication fraction, whereas lane b contained 0.6 ,ug of the dsDNA-cellulose reactions (40 ,ul) as described in Materials and Methods with 0.45 Jg fraction. The numbers above the lanes indicate the gradient fractions of T antigen. Replication products were analyzed by 1.2% alkaline used in the assay. DNA size markers (in nt) are indicated at the left. agarose gel electrophoresis as described (14) (A) and by quantitation (C) Exonuclease III assay ofthe gradient fractions. Lane a contained of acid-insoluble material formed (B). DNA size markers (in nt) are 0.3 jg ofthe dsDNA-cellulose fraction. The numbers above the lanes indicated at the left in A. indicate the gradient fractions (0.5 ,.Il) assayed. Downloaded by guest on September 29, 2021 Biochemistry: Lee et al. Proc. Natl. Acad. Sci. USA 86 (1989) 4879 mM NaCl (5 ml, 3.7 mg of protein, 1 x 104 units; specific preparation was homogeneous and fully active, the PCNA activity, 2702 units/mg ofprotein) and then chromatographed activation was evident at molar concentrations that were 10- on a double-stranded (ds) DNA-cellulose column (1.2 x 1.8 to 100-fold lower than the molar concentration of the inhib- cm; 2 ml) equilibrated with buffer A containing 0.2 M NaCl. itor. This column was eluted successively with 10 ml of buffer A The Elongation Inhibitor Is a 120-kDa Protein. The dsDNA- containing 0.2, 0.4, 0.6, and 0.8 M NaCl. The inhibitor eluted cellulose fraction was rechromatographed on a dsDNA- with 0.4 M NaCl. This fraction was dialyzed against buffer A cellulose column and eluted with a linear salt gradient (Fig. containing 0.025 M NaCl (3 ml, 1.77 mg of protein, 7.5 x 103 2). Fractions were analyzed by SDS/PAGE (Fig. 2A) and units; specific activity, 4237 units/mg of protein) and then assayed for their ability to block SV40 replication (Fig. 2B) centrifuged through a glycerol gradient. The dsDNA- and inhibit the action of exonuclease III (Fig. 2C). The peak cellulose fraction (0.2 ml, 0.12 mg of protein) was layered of inhibitory activity for the two activities coincided with the onto a 5-ml 15-35% glycerol gradient containing buffer A plus presence of a 120-kDa protein band. No other protein coin- 0.25 M NaCl and centrifuged at 45,000 x g for 24 hr at 4(C. cided with the inhibitory activity across the gradient. Thirty fractions (0.17 ml each) were collected and assayed for Gel-filtration chromatography was performed to define the the inhibitor. The inhibitor activity peaked in fractions 16-20, native size and structure of the elongation inhibitor. The which were then pooled (0.85 ml, 40 Ag ofprotein, 283 units; 120-kDa elongation inhibitor was eluted from a dsDNA- specific activity, 7075 units/mg of protein). If the entire cellulose column with a NaCl linear gradient and concen- dsDNA-cellulose fraction had been carried through this step, trated by ammonium sulfate precipitation (60%). The sample a total of4.2 x 103 units would have been recovered. The final (0.22 mg) was chromatographed on Sephacryl S200-SF (0.7 x fraction of the elongation inhibitor was purified approxi- 50 cm; Pharmacia) at 40C in buffer containing 25 Tris HCl (pH mately 17-fold over the 0.7 M PC starting fraction with a yield 7.5), 0.25 M NaCl, 1 mM dithiothreitol, 1 mM EDTA, 0.1 mM of about 10%o. phenylmethylsulfonyl fluoride, 0.2 ,ug of antipain per ml, 0.1 Protein Assays. Protein concentrations were determined by gg of leupeptin per ml, and 0.02% sodium azide. Active using the Bio-Rad protein assay reagent. SDS/PAGE elec- fractions containing the 120-kDa elongation inhibitor eluted trophoresis of protein fractions was carried out as described in a fraction corresponding to a Stokes radius of 59.5 A. by Laemmli (16), and protein bands were visualized using the Based on a sedimentation coefficient of 4.2S (14), the native ICN Rapid-Ag-Stain kit. molecular mass of the elongation inhibitor is 138 kDa as determined by the Siegel and Monty equation (17). The 138-kDa fraction was active both as an elongation inhibitor in RESULTS the SV40 DNA replication assay in vitro and in blocking the Properties of the Elongation Inhibitor. With increasing action of exonuclease III. On SDS/PAGE analysis, the concentrations of the inhibitor, SV40 ori+ DNA synthesis 138-kDa fraction migrated as a 120-kDa protein (data not decreased and shorter DNA chains accumulated (Fig. 1). shown). When PCNA was added, the size of the DNA products as Replication Elongation Inhibitor Blocks Enzymatic Reac- well as the amount of DNA synthesis increased (Fig. 1). tions That Act at the Ends of DNA. The dsDNA-cellulose Relatively low (nanogram) concentrations of PCNA could fraction was separated by glycerol gradient centrifugation, reverse the effects of large (microgram) amounts of the and the fractions were assayed for their ability to inhibit the inhibitor, suggesting that PCNA may act catalytically. In- replication of SV40 ori+ DNA in the presence and absence of creasing concentrations ofthe inhibitor appeared to decrease PCNA. The inhibitor activity peaked in fraction 18 of the the length of DNA chains proportionally. In contrast, the glycerol gradient (Fig. 3). The dsDNA-cellulose inhibitor synthesis of nearly full-length chains was stimulated by low activity, which was loaded on to the gradient, was reversed (10 ng) concentrations of PCNA, which barely increased by PCNA (Fig. 3, lanes B and C). In contrast, the inhibitory nucleotide incorporation. This marked increase in length is activity offraction 18 was only marginally reversed by PCNA characteristic of a processive system. Assuming that each (Fig. 3, lanes 16-20). Glycerol gradient fractions 10 and 12 A B C 2 4 6 8 10 12 14 16 18 20 22 24 26 r- A r ln - r- r--'~ r-- r f , PCNA -

t 10 * to w I * 2480 - FIG. 3. Sedimentation of the elongation inhibitor by glycerol gradient centrifugation. The ds- DNA-cellulose fraction was subjected to glycerol gradient centrifugation. Gradient fractions (2 ,ul each) were assayed for their effects on the SV40 replication. Products were subjected to 1.2% alkaline agarose gel electrophoresis. Lane A contained no elongation inhibitor fraction, whereas 310 - lanes B and C each contain 0.6 ,ug n. t. of the 0.4 M NaCl dsDNA-cellulose fraction of the elongation inhibitor. Where indicated, 0.2 ,g of purified PCNA was added. The numbers on the top of the figure indicate the glycerol gradient fractions used in the assays. Downloaded by guest on September 29, 2021 4880 Biochemistry: Lee et al. Proc. Natl. Acad. Sci. USA 86 (1989) stimulated DNA synthesis in the presence ofPCNA, and this activity was shown to be a property of the 120-kDa protein. observation is discussed in more detail below. The glycerol gradient fraction of the inhibitor was subjected We examined the effect of the inhibitor on a number of to SDS/PAGE, transferred to nitrocellulose, and probed with enzymatic activities that occur at the ends of DNA. The dA4wo0[5'-32P]dT5o. The only labeled band detected corre- inhibitor blocked the action of exonuclease III (Fig. 4A), the sponded to a protein of 120 kDa (data not shown). elongation ofa primed DNA template by DNA polymerase a, The elongation inhibitor is a basic protein with an isoelec- the action of DNA ligase (using dA4000-[5'-32P]dT50 as sub- tric point estimated to be pH 9.0 or greater using the PhastGel strate), and the HeLa 5' to 3' exonuclease [the enzyme that isoelectric focusing 3-9 system (Pharmacia). removes RNA primers in the SV40 replication reaction (11)] The inhibitor was resistant to N-ethylmaleimide and sen- (Fig. 4B). When these enzyme activities were used to monitor sitive to heat (50o inactivated after 55 sec at 500C). Both the the sedimentation of the elongation inhibitor in the glycerol binding ofdA400-[5_-32P]dT50 and the inhibition of replication gradient, the profile ofinhibitory activity was the same as that were heat inactivated at the same rate. The complex formed observed with the SV40 replication system (Fig. 3). with the inhibitor and the singly primed 4X174 DNA was Other Properties ofthe Elongation Inhibitor. The elongation stable to gel filtration and retained its ability to block both inhibitor (glycerol gradient fractions 16-20) bound both ss exonuclease III and DNA polymerase a-catalyzed reactions. linear and ss circular DNA efficiently. The DNA-binding The addition of HeLa SSB or ATP did not affect the binding of the inhibitor to primed DNA (data not shown). A Glycerol Gradient Fraction Resolution of the Elongation Inhibitor from a PCNA- Mediated Activator. The inhibitor activity in fractions 16-20 A B C 2 4 6 8 10 12 14 16 18 20 22 24 26 28 from the glycerol gradient was only marginally reversed by 32 mer- 0 * *S 0. PCNA in replication reactions containing T antigen, the 0.3 M NaCl PC fraction, topo I, and the HeLa SSB (Fig. 3). In contrast, fractions 10-12 from the gradient stimulated the formation oflonger DNA products in the presence of PCNA. dAMP Therefore the influence offractions 10-12 in overcoming the effect of the fraction 16-20 inhibitor was examined (Fig. 5). As shown in Fig. 5, lane 6, the addition offractions 10-12 and PCNA reversed the effect of the fraction 16-20 inhibitor, resulting in the synthesis of long DNA chains. These results indicate that an additional protein required for the PCNA- B dependent reversal of the block in SV40 replication was 100100(^ \;o~~ separated from the elongation inhibitor by glycerol gradient 011 ~~~ADNA ic 0 f/ 1 2 3 4 5 6 7 >% P a 0 6$0K 2480- Ilt!--zIc-.

4 8 12 16 20 24 28 30 Glycerol Gradient Fractions FIG. 4. Influence of glycerol gradient fractions on the activity of exonuclease 111, 5' to 3' exonuclease, DNA ligase, and the elongation reaction catalyzed by DNA polymerase a. (A) Assay ofexonuclease 310 - III activity. The assay was carried out with 0.4 A1 of each glycerol gradient fraction. Lanes: A, no protein was added; B, 0.04 unit of n.t. exonuclease III; C, 0.3 ,g of dsDNA-cellulose fraction plus exonuclease III. All other lanes contained exonuclease III. (B) The 5' to 3' exonuclease (5'-3' Exo) was measured in reaction mixtures (50 ,ul) as previously described (11). Reactions with 2 Al of the gradient 0.3 M PC + + + + + ++ fractions were preincubated for 5 min at 25°C, and then 2 units of the 5' to 3' exonuclease was added. Incubations were for 15 min at 37°C PCNA followed by the addition of 20 ,ul of bovine serum albumin (5 mg/ml) 0.4MdsDNA and 160 ,ul of ethanol. The mixtures were stored at -80°C for 30 min G. G. (#10-12) .++ and then centrifuged, and the supernatant was measured for the G.G. (#16-20) ethanol-soluble 32p. The 100o value was equivalent to the release of dNMP incorp. 51 12 46 26 27 39 18 50 fmol of 32p. The assay for DNA ligase activity was carried out in (pmol) reaction mixtures (50 ,ul) as previously described (11). After 5 min of preincubation with 2 p.1 ofgradient fraction at 25°C, DNA ligase (0.04 FIG. 5. Resolution of the PCNA-dependent activator I from the unit) was added, and the reaction mixtures were incubated at 37°C for elongation inhibitor by glycerol gradient centrifugation. Glycerol 30 min. The DNA ligase activity was determined as previously gradient fractions 10-12 (Fig. 3) were pooled as were glycerol described (11). The 100%o value in this assay was equal to 15 fmol of gradient fractions 16-20. The pools were then dialyzed against buffer 32p rendered resistant to alkaline phosphatase. Each glycerol gradi- A containing 20% glycerol and 25 mM NaCl. The influence of the ent fraction (3 ,ul) was used for the DNA polymerase a (Pol a) pooled glycerol gradient fractions and PCNA on the replication of elongation assay; the incorporation of 12 pmol of dTMP was equiv- SV40 DNA was carried out; the amount of dNMP incorporated was alent to 100%. measured on aliquots of each reaction mixture and is indicated. Downloaded by guest on September 29, 2021 Biochemistry: Lee et al. Proc. Natl. Acad. Sci. USA 86 (1989) 4881 centrifugation. This activator fraction has been named acti- were formed by a crude receptor fraction, which contained vator fraction I. PCNA as well as polymerase-primase, HeLa SSB, and other As previously shown (14), the dsDNA-cellulose fraction of factors. These observations are analogous to our findings the inhibitor (prior to the glycerol gradient step) blocked with reactions containing inhibitor but lacking activators. replication reactions in which immunopurified DNA poly- The activation of primer ends described here has some merase a-primase complex was used instead of the 0.3 M similarities to DNA synthesis catalyzed by the DNA poly- NaCl PC fraction. When the immunopurified polymerase a- merase III system ofE. coli. In this system, the y-8 complex primase complex was used, the inhibitor activity could not be catalytically transfers the dnaN gene product (,B subunit or reversed by the addition of PCNA and the glycerol gradient factor I) to the 3'-OH end ofa primed template in the presence activator fraction (data not shown). This suggests that an- ofATP. A complex that includes at least the primed template other activator, which we have called activator fraction II, is and the dnaN gene product is then utilized by DNA poly- present in the 0.3 M NaCl PC fraction. merase III in the elongation reaction, which results in the processive growth of chains at high rates (19-21). In the absence ofthe dnaN gene product, DNA polymerase III acts DISCUSSION distributively. It is possible that PCNA, the inhibitor, and the In this report, we have further characterized an elongation elongation activators act in a manner analogous to the y-8 inhibitor that acts at the end of DNA chains. This protein complex and the dnaN protein. interferes with enzymatic reactions that occur at 3' or 5' ends Further investigation of the mechanism by which the ofDNA, such as the elongation ofa primed template by DNA activators and PCNA interact with the elongation inhibitor polymerase a, the joining reaction catalyzed by DNA ligase, will be required to understand the control ofDNA elongation and both 5' to 3' and 3' to 5' exonuclease reactions. The in the SV40 replication system. unusual property of the elongation inhibitor is that its action Note Added in Proof. We have identified the elongation inhibitor as in the SV40 system is reversed by a combination of PCNA poly(ADP-ribose) polymerase and activator II as PCNA-dependent and other protein activators. The PCNA-dependent reversal DNA polymerase 8. can also be demonstrated directly either in reactions catalyzed by DNA polymerase a or exonuclease III by using We thank Ms. B. Phillips and Ms. N. Belgado for their assistance. primed 4X174 DNA as a substrate for both systems (data not This work was supported by Grant GM-34559 from the National shown). Institutes of Health. A.D.K. is a Leukemia Society of America SV40 DNA replication by relatively crude fractions in the Fellow. absence of PCNA is primarily restricted to the formation of small DNA fragments from lagging-strand DNA synthesis 1. DePamphilis, M. L. & Wasserman, P. M. (1982) in Organiza- (13). We have made similar observations with reaction mix- tion and Replication of Viral DNA, ed. Kaplan, A. S. (CRC, tures lacking PCNA but containing the 0.3 M NaCl PC Boca Raton, FL), pp. 37-144. 2. Varshavsky, A., Sundin, O., Ozkaynak, E., Pan, R., Solomon, fraction, purified HeLa SSB, topo I (or topo II), T antigen, M. & Snapka, R. (1983) in Mechanisms of DNA Replication and the inhibitor (data not presented). At present, it is not and Recombination, ed. Cozzarelli, N. R. (Liss, New York), clear why only lagging-strand synthesis is observed in the pp. 463-494. absence of PCNA, since the elongation inhibitor does not 3. Li, J. J. & Kelly, T. (1984) Proc. Natl. Acad. Sci. USA 81, appear to discriminate between leading- and lagging-strand 6973-6977. DNA synthesis. In the SV40 DNA replication system, we 4. Stillman, B. W. & Gluzman, Y. (1985) Mol. Cell. Biol. 5, 2051- have shown that small DNA fragments synthesized during a 2060. pulse period (in reactions containing 0.3 M NaCl PC fraction, 5. Wobbe, C. R., Dean, F., Weissbach, L. & Hurwitz, J. (1985) HeLa SSB, topo I, T antigen, and the inhibitor but no PCNA) Proc. Natl. Acad. Sci. USA 82, 5710-5714. 6. Murakami, Y., Wobbe, C. R., Weissbach, L., Dean, F. & can be chased into long DNA products by the addition of Hurwitz, J. (1986) Proc. Natl. Acad. Sci. USA 83, 2869-2873. PCNA and the activator fraction I (data not shown). This 7. Yang, L., Wold, M. S., Li, J. J., Kelly, T. & Liu, L. F. (1987) suggests that the reversal of the action ofthe inhibitor allows Proc. Natl. Acad. Sci. USA 84, 950-954. DNA ligase and the 5' to 3' exonuclease to synthesize longer 8. Wobbe, C. R., Weissbach, L., Borowiec, J. A., Dean, F. B., DNA products from the short lagging-strand products. In the Murakami, Y., Bullock, P. & Hurwitz, J. (1987) Proc. Natl. SV40 replication system reconstituted with purified compo- Acad. Sci. USA 84, 1834-1838. nents, the 5' to 3' exonuclease and DNA ligase are essential 9. Wold, M. S. & Kelly, T. (1988) Proc. Natl. Acad. Sci. USA 85, for the synthesis offull-length and circular DNA products. In 2523-2527. their absence both small Okazaki size (200 nt) and 10. Fairman, M. P. & Stillman, B. (1988) EMBO J. 7, 1211-1218. half-length 11. Ishimi, Y., Claude, A., Bullock, P. & Hurwitz, J. (1988) J. Biol. (1700 nt) products were formed using pSV01A&EP DNA as the Chem. 263, 19723-19733. template (11). 12. Prelich, G., Kostura, M., Marshak, D. R., Mathews, M. B. & Low concentrations of PCNA in the presence of the Stillman, B. (1987) Nature (London) 326, 471-475. activator fractions reversed the effects of relatively large 13. Prelich, G. & Stillman, B. (1988) Cell 53, 117-126. amounts of the inhibitor. The amount of elongation activator 14. Lee, S.-H., Ishimi, Y., Kenny, M. K., Bullock, P., Dean, F. & protein I isolated by glycerol gradient centrifugation was at Hurwitz, J. (1988) Proc. Natl. Acad. Sci. USA 85, 9469-9473. least 40-fold lower than the amount of elongation inhibitor 15. Lee, M. & Marians, K. (1987) Proc. Natl. Acad. Sci. USA 84, isolated (Fig. 3). These observations suggest that the activa- 8345-8349. tor may act catalytically in reversing the effect of the inhib- 16. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 17. Siegel, L. M. & Monty, K. J. (1966) Biochim. Biophys. Acta itor. 112, 346-362. Recently, Tsurimoto and Stillman (18) reported the isola- 18. Tsurimoto, T. & Stillman, B. (1989) Mol. Cell. Biol. 9, 609-619. tion ofa complex (containing four polypeptides of37, 41, 100, 19. Wickner, S. (1976) Proc. Natl. Acad. Sci. USA 73, 3511-3515. and 140 kDa) required for the synthesis of long DNA prod- 20. O'Donnell, M. E. (1987) J. Biol. Chem. 262, 16558-16565. ucts. In its absence, short, lagging-strand DNA fragments 21. Maki, S. & Kornberg, A. (1988) J. Biol. Chem. 263, 6561-6569. Downloaded by guest on September 29, 2021