VIROLOGY
182,
376-381
(1991)
Characterization of 3’40-5 with Epstein-Barr
Exonuclease Activity Associated Virus DNA Polymerase
TATSUYA
TSURUMI
Laboratory of Virology, Research Institute for Disease Mechanism and Control, Nagoya University School of Medicine, Showa-ku, Nagoya 466, Japan Received November
6, 1990; accepted January 24, 199 1
Epstein-Barr virus (EBV) DNA polymerase mediates viral DNA replication during the lytic phase of the EB virus life cycle. In order to characterize its enzymatic activities EBV DNA polymerase was purified more than 1200-fold from chemically induced 895-8 cells. One polypeptide with molecular weight of 110,000 corresponded to the predicted EBV DNA polymerase, whereas the other polypeptides did not. A 3’-to-5’ exonuclease activity was copurified with the EBV DNA polymerase through the course of the purification. Unlike HSV DNA polymerase, 5’-to-3 exonuclease activity was not associated with the EBV DNA polymerase on the final step chromatography of single-stranded DNA agarose column. The associated 3’-to-5’ exonuclease activity was stimulated by ammonium sulfate like the polymerase activity. It exhibited DNA-dependent nucleotide turnover activity and preferentially excised a terminal mismatched nucleotide on hybridized polynucleotides compared to the correctly paired substrate, indicating that the 3’-to-5 exonuclease may play a role in proofreading in the polymerization process. 0 1991 Academic Press, Inc.
Epstein-Barr virus (EBV) is a human B lymphotropic herpesvirus which is a causative agent of infectious mononucleosis and is known to be closely associated with African Burkitt’s lymphoma and nasopharyngeal carcinoma (1). EBV infects and immortalizes B lymphocytes, in which only a limited number of viral genes are expressed and the EBV genome is maintained as a circular plasmid molecule by using host DNA polymerases. OriP, an origin of EBV, mediates this type of replication during the latent phase of EBV life cycle (2). In a small fraction of the immortalized B cells most of viral genes are expressed to yield the lytic phase of the life cycle. Chemical agents such as 12-0-tetradecanoyl phorbol-13-acetate (TPA) or sodium n-butyrate can induce the lytic phase itn some cell lines latently infected with EBV (3). During the lytic phase of viral life cycle EBV DNA is amplified 100 to 1000 times. The replication product is a head-to-tail concatemer, which is synthesized by EBV DNA polymerase presumably via rolling circle mechanism initiated from oriLyt (4). Thus, EBV DNA polymerase acts as a key enzyme during the lytic phase of EBV DNA replication. The complete nucleotide sequence of the B95-8 strain of EB virus was determined by Baer et a/. and likely protein-coding regions were identified (5). The genome of EBV B95-8 consists of 172,281 bp and encodes 84 proteins. EBV DNA polymerase gene was identified and localized on the BALF5 open reading frame of the EBV genome (5, 6). Estimated molecular weight from the open reading frame is 1 13,400 Da. 0042-6822/91
$3.00
Copyright 0 1991 by Academic Press. Inc. All rights of reproduction !n any form reserved
EBV-induced DNA polymerase has been partially purified (6- 12) and characterized with respect to reaction optima, sensitivities to various inhibitors, and template specificities. However, the enzymatic characterization of EBV DNA polymerase is not enough to understand the DNA synthesis process, especially with respect to its associated nuclease activity. In this study I have purified EBV DNA polymerase from B95-8 cells treated with both TPA and sodium n-butyrate and characterized EBV DNA polymerase with respect to its associated exonuclease activity in detail. B95-8 cells were treated with 100 rig/ml TPA and 5 mM sodium n-butyrate to induce the lytic phase of EBV DNA replication. Salt-stimulated DNA polymerase activity was induced after addition of the chemical agents. The DNA polymerase was extracted from the nuclei of induced B95-8 cells. The salt-stimulated DNA polymerase activity was detected neither in the cytosol fraction nor in the low-salt extract of the nuclei, but fractionated into 0.4 M NaCl extract of the nuclei. This first step purification was very useful to separate the EBV DNA polymerase from host DNA polymerase (Y, which was fractionated into cytosol, and to increase purification (Table 1). Figure 1 shows the purification of the viral DNA polymerase by sequential chromatography on phosphocellulose (A), heparin agarose (B), and single-stranded DNA agarose (C). On the phosphocellulose column chromatography EBV DNA polymerase eluted at 0.35 n/l NaCI. A DNase eluted at 0.25 M NaCI. The DNase could be EBV-induced DNase because its 376
SHORT COMMUNICATIONS
377
TABLE 1 SUMMARYOFPURIFICATIONOF EBV DNA POLYMERASE
Fraction I II Ill IV V
Step Total cell extract (Cytosol + nuclei) Medium salt nuclear extract Phosphocellulose Heparin agarose ssDNA agarose
Total protein (w)
Total activity (units)
Specific activity (units/mg)
Overall yield (%)
Purification (fold)
238
568,000
2,386
100
1
20
492,800
24,640
87
19
1.37 0.494 0.052
414,720 268,800 157,500
302,715 544,130 3,028,846
73 47 28
127 228 1269
Note. One unit is defined as 1 pmol of dNMP incorporation
for 60 min.
activity was 8 times as high as the activity observed in case of uninduced 695-8 cell extracts. On heparin agarose column chromatography EBV DNA polymerase eluted at 0.42 M NaCl and was completely separated from the contaminating DNase activity, which eluted at 0.35 M NaCI. However, some nuclease activity was detected along with the polymerase peak (Fig. 1B). On the single-stranded DNA agarose column, the DNA polymerase eluted at 0.35 M NaCI. The summary of the purification of EBV DNA polymerase was given in Table 1. The EBV DNA polymerase was purified 1269-fold from the whole cell extract with a yield of 28%. The peak fraction of the EBV DNA polymerase on the single-stranded DNA agarose column chromatography contained three protein bands, of which a 110. kDa polypeptide coincided with the peak of the polymerase activity (Fig. 1 D). The remaining two proteins (50 and 48 kDa) eluted with peaks at 0.25 M NaCl on the single-stranded DNA agarose column, of which fractions were lacking the DNA polymerase activity. Thus, only the 1 10-kDa protein corresponded to the predicted EBV DNA polymerase. The peak fractions of the EBV DNA polymerase were devoid of detectable primase and ATPase activities. The enzyme exhibited a broad optimum activity between pHs 6.8 and 9.2. The DNA polymerase activity was stimulated more than 2-fold by 75 to 100 mM ammonium sulfate and was resistant even at 150 mM ammonium sulfate, while DNA polymerase (Ywas very sensitive to ammonium sulfate (50% inhibition at 30 mM). The EBV DNA polymerase utilized poly(dC). oligo(dG) 29-fold more efficiently than activated DNA, but the RNA template-DNA primer, poly(rA) . oligo(dT), could not be utilized at all. Further, the polymerase activity was inhibited by 70% at 10 pg/ml phosphonoacetic acid and by 80% at 0.5 mM A/-ethylmaleimide. Aphidicolin was also inhibitory to the EBV DNA polymerase as well as to DNA polymerase CL 50% inhibition was observed at 2 pg/ml. The above enzymatic
properties were specific for EBV DNA polymerase as was reported previously (6, 12, 13) and clearly distinguishes the viral enzyme from the host DNA polymerase LY,6, y, or 6 (6, 14, 75). Recent studies have shown that 3’-to-5’exonuclease activity is associated with other herpesvirus DNA polymerases ( 16- 18). To analyze the nuclease activity associated with the EBV DNA polymerase in detail, 3’-to5’ exonuclease activity on fractions of the phosphocellulose, heparin agarose, and single-stranded DNA agarose column chromatography were assayed using 3’-terminal labeled poly(dA-dT) as substrate in the presence of 80 mM ammonium sulfate. As shown in Fig. 1, the 3’-to-5’ exonuclease activity was found to cochromatograph with the EBV DNA polymerase through the course of the purification. Recently, it has been reported that herpes simplex virus type 1 DNA polymerase contains not only 3’-to-5’ exonuclease but also 5’-to-3’ exonuclease (19). To determine this possibility for the EBV DNA polymerase, 5’-to-3’ exonuclease (Fig. 1C) was assayed by using [5’-32P]poly(dA),,, . poly(dT),ooo as substrate. However, 5’-to-3’exonuclease eluted ahead of the EBV DNA polymerase and was not associated with the EBV DNA polymerase fractions on the final step of chromatography on single-stranded DNA agarose column. The 3’-to-5’ exonuclease activity was assayed by using 3’-terminally labeled poly(dA-dT) alternating copolymer as substrate. The reaction product was dTMP determined by PEI thin-layer chromatography (data not shown), indicating that the EBV DNA polymerase hydrolyzed 3’-terminal nucleotide. The effect of adding increasing amounts of ammonium sulfate was determined (Fig. 2). The polymerase associated 3’-to-5’exonuclease was stimulated by 25 to 100 mM ammonium sulfate. Even at 150 mM ammonium sulfate 80% of the activity remained. This pattern of stimulation by ammonium sulfate was analogous to that observed with the EBV DNA polymerase activity. On the other hand, the
SHORT COMMUNICATIONS
378
(u!w OZ/~.O t x loud) ~NI=
aseJawA@
t/NO MEI
l
(c-or x WW I+ aiqnlos p!zv :aseNa
,-yy=-=j-jB
o
SHORT COMMUNICATIONS
nuclease activity of EBV-induced DNase, which eluted ahead of EBV DNA polymerase on the phosphocellulose column chromatography in the same assay system, was more sensitive to the salt (Fig. 2). Fifty percent inhibition was observed at 60 mA4 ammonium sulfate. Thus, the 3’-to-5’ exonuclease activity associated with the viral DNA polymerase was totally different from that of EBV-induced DNase. The other properties of the 3’-to-5’ exonuclease are summarized in Table 2. Mg2+ was essential for the reaction and could be replaced by 1 mM Mn2+. The exonuclease activity was very sensitive to 3 mn/l NEM. However, aphidicolin and phosphonoacetic acid had almost no inhibitory effect on the exonuclease activity at a concentration of 15 pg/ml and 13 pg/ml, respectively, at which EBV DNA polymerase activitywas inhibited by 809/o. Brutlag and Kornberg (20) demonstrated the proofreading activity of E. co/i polymerase I by measuring the removal of a terminal mismatched nucleotide. To determine whether the 3’-to-5 exonuclease associated with EBV DNA polymerase has proofreading activity, removal of 3’-terminal labeled nucleotide from
pob(dA)~(dT)20[3Hl(dT),., or wM@J - (dT)2,[3WG),.2 was examined (Fig. 3). The paired 3’ terminus was stable to the 3’-to-5’ exonuclease in the presence of dlTP, but excision of the correctly paired terminus in
379
the absence of dlTP was observed (Fig. 3B). This is most likely due to the addition of nucleotide to the paired 3’terminus by the polymerase. In contrast, even in the presence of dTTP the 3’-to-5’ exonuclease rapidly removed the unpaired 3’-terminal nucleotide (Fig. 3A). The 3’-to-5’ exonuclease hydrolyzed an unpaired 3’terminal nucleotide at about twice the rate at which the paired 3’-terminal nucleotide was hydrolyzed. These observations suggest that excision of the mismatched terminal nucleotide is necessary before elongation can proceed. DNA-dependent nucleotide turnover is a characteristic of DNA polymerases possessing 3’-to-5’exonucleases (21). EBV DNA polymerase exhibited nucleotide turnover from deoxynucleotide triphosphate to monophosphate form when poly(dC) . oligo(dG) or activated DNA was used as template primer (Table 3). Both turnover assays were done in the presence of controls, which were identical assays performed in the absence of the templates. No deoxynucleotide monophosphate was produced in such control experiments (data not shown), demonstrating that this reaction was template dependent. The dGMP or dTMP is presumably formed upon hydrolysis of paired 3’termini during DNA synthesis as observed previously for the HSV DNA polymerase (18) and for other DNA polymerases possessing 3’90-5 exonuclease activities (2 I).
FIG. 1. Purification of EBV DNA polymerase. The 895-8 cells were seeded at 1O6 cells/ml in culture medium. After 30 hr of cultivation, the chemical inducers were added at final concentrations of 100 rig/ml for TPA and 5 mM for sodium n-butyrate. The cells were harvested at 48 hr postinduction, washed, and then suspended in hypotonic lysis buffer (20 mM Na-HEPES, pH 7.2, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF, 10 pg/ml pepstatin A, and 10 pg/ml leupeptin) for 30 min at 0”. The nuclei were prepared by Dounce homogenization and finally suspended in the stock buffer (20 mM Na-HEPES, pH 7.2, 1 mM EDTA, 1 mM EGTA. 1 mM DTT, 1 mM PMSF, 10 pg/ml pepstatin A, 10 pglml leupeptin, and 10% sucrose) and stored at -80”. The stored nuclear preparations were thawed and collected by low-speed centrifugation. The pellet was suspended in buffer A (20 mM Tris-HCI, pH 7.2, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF, 10 pg/ml pepstatin A, 10 pg/ml leupeptin) containing 0.4 M NaCl and centrifuged at 20,000 rpm in a Hitachi SW28 rotor for 1 hr at 4”. The supernatant was dialyzed against buffer A containing 0.1 M NaCl overnight (Fraction II). (A) Fraction II was loaded onto a phosphocellulose (P 1 1, Whatman BioSystems Ltd.) column (5 ml bed volume) equilibrated with buffer A containing 0.1 M NaCI. The phosphocellulose column was washed with 25 ml buffer A containing 0.1 M NaCl and eluted with a 60.ml linear gradient from 0.1 to 0.8 M NaCl In bufferA. Peak fractions were pooled and dialyzed against 100 vol of buffer A containing 0.1 M NaCl (Fraction Ill). (B) Fraction Ill was applied to a column (1.5 ml bed volume) of heparin agarose equilibrated with buffer A containing 0.1 M NaCI. The column was washed and eluted with a 30.ml linear gradient from 0.1 to 1 .O M NaCl in bufferA. Active fractions were pooled and dialyzed against 100 vol of buffer A containing 0.1 M NaCl (Fraction IV). (C) Fraction IV was applied to a column (1 ml bed volume) of single-stranded DNA agarose equilibrated with bufferA containing 0.1 M NaCI. The column was washed and eluted with a 24-ml linear gradient from 0.1 M to 0.8 M. EBV DNA polymerase (0) was assayed in reaction mixture (25 pl) that contained 50 mM Tris-HCI, pH 8.4, 80 mM ammonium sulfate, 10% glycerol, 6 mM MgCI,, 100 pg/ml bovine serum albumin, 80 pg/ml activated calf thymus DNA, 1 mM DTT, 25 PM each of dATP, dGTP, and dCTP, and 10 FM [3H]lTP (1800 cpm/pmol). The reaction was started by the addition of 2 ~1of enzyme fraction, After 20 mm at 35”, TCA precrpitable counts were determined. Assay for DNase activity (0) was done with E. co/i DNA labeled with [3H]thymidine to a specific actrvity of 38,000 cpm/pg. Each assay of 25 ~1 contained 1 pg of labeled DNA, 50 m/L?Tris-HCI, pH 8.4, 10% glycerol, 6 mM MgCI,, 100 rg/ml BSA, 1 mM DTT. The reaction was started by the addition of 4 ~1 of enzyme fraction, After 30 min at 35”, released nucleotides were determined (19). 3’-to-5’ exonuclease activity (0) was assayed as follows, Reaction mixture (25 ~1) contained 40 mM Tris-HCI, pH 8.4, 80 m/L? ammonium sulfate, 6 mM MgCI,, 100 pglml BSA, 1 mM DTT, and 2 pg/ml [3’-3H]poly(dApdT) (1.5 X 1O5 cpmlpg). Four microliters of enzyme fraction was added to start the reaction. After 10 min, released nucleotides were determined (76). 5’.to-3’ exonuclease activity (A) was assayed in reaction mixture (25 ~1) containing 40 mM Tris-HCI. pH 8.4, 3 mM MgCI,, 100 @g/ml BSA, 1 mM DlT, and 10 pM [5’-32P]poly(dA)~poly(dT),,,, (445 cpmipmol terminal nucleotide). Four microliters of each fraction was added to start the reaction. After 20 mm, released nucleotides was determined (19). (D) Silver staining of SDS-PAGE of EBV DNA polymerase. Aliquots of the peak fraction (26) of the srngle-stranded DNA agarose column subjected to 8% SDS-polyacrylamide gel electrophoresis and stained with srlver. DF IS the dye front. The positions of the molecular weight standards (M, X 1Om3)are indicated by arrows.
380
SHORT COMMUNICATIONS B
--e-.-. -T-
\
T’
-dd-di-dA-
0 \
OlO
I--
t 0-
20
Time ( min )
FIG. 3. The 3’.to-5’ exonuclease activity of EBV DNA polymerase on terminally labeled nucleotides. The 3’-to-5’ exonuclease activity on the 3’.unpaired (A) and -paired (B) substrates were assayed in the presence (0) or absence (0) of dTTP. The reaction contained
0 0
50
100
(NH+S04
150
(mM)
FIG. 2. Effect of ammonium sulfate on the 3’.to-5’ exonuclease activity associated with EBV DNA polymerase and that of EBV-induced DNase. The r-to-5 exonuclease activity of EBV DNA polymerase (0) and that of EBV-induced DNase (0) were assayed using [3’-3H]poly(dA-dT) as substrate as described in the legend to Fig. 1 except that ammonium sulfate concentration was varied as indicated. The peak fractions of the DNase which eluted ahead of EBV DNA polymerase on the phosphocellulose column chromatograph (Fig. 1A) were applied to heparin agarose column chromatography. The peak fraction of the DNase on the heparin agarose column chromatograph was used as the sample for the EBV-induced DNase. Two microliters of enzyme fraction was applied to start the reaction. The activity is expressed as percentage of the activity obtained in the absence of ammonium sulfate.
The peak fraction of EBV DNA polymerase on the single-stranded DNA agarose column chromatograph contained three protein bands, of which the 1 lo-kDa polypeptide corresponded to the polymerase activity and the 3’-to-5’ exonuclease activity. It was in agreement with estimated size from the open reading frame
TABLE 2 CHARACTERISTICSOF THE 3’-TO-~’ EXONUCLEASEACTIVITV ASSOCIATED WITH EBV DNA POLYMERASE
Omission or addition None Omit Omit Omit Add Add
(complete) Mgz+ Mg*+ plus Mn2+ (1 mM) DTr plus N-ethylmaleimide (3 mM) aphidicolin (15 pglml) phosphonoacetic acid (13 pglml)
40
Time ( min )
dTMP released (% of maximum) 100 2 98 4 88 89
Note. The 3’-to-5’ exonuclease activity of the peak fraction (26) of EBV DNA polymerase on the single-stranded DNA agarose column chromatography was measured using [Y-3H]poly(dA-dT) as described in the legend to Fig. 1.
poWA) *kWd3WdG), 2 (4 or poWA) *W)20-[3H1(dT)1.4 (W as substrate. Excision of a terminal mismatch was examined as described by Brutlag and Kornberg (79). Reaction mixture (25 pl) contained 40 mM Tris-HCI, pH 8.4, 1 mM DTT, 6 mM MgCI,, 100 pg/ml BSA, 80 mM ammonium sulfate, and 0.1 pg of template primer, which was either poly(dA) . (dT)2,-[3H]dT,., (704 cpm/pmol terminal nucleotide) or poly(dA). dT,JH]dG,,, (1059 cpm/pmol terminal nucleotide). When present, dTTP was 100 pM each. Reactions were carried out at 30” to reduce partial denaturation of the primer template. The exonuclease activity was measured by the reduction in DE 81 absorbed radioactivity (26).
of BALF5. It has recently found that herpes simplex virus type 1 DNA polymerase is a heterodimer complex which consists of the products of UL30 (Pol) and UL42 (Pol accessory protein) genes (19). In the case of Epstein-Barr virus, it has been reported that BMRF 1 gene product has an ability to stimulate the EBV DNA polymerase activity (10, 22, 23). Moreover, BMRF 1 gene product possesses some amino acid sequence homology with HSV-1 UL42 gene product (T. Tsurumi, unpublished data). The estimated size of BMRF 1 gene product is 43K (5). The peak fraction of EBV DNA polymerase on the single-stranded DNA agarose column chromatography contained two other polypeptides, 50 and 48 kDa, which did not correlate with the polymerase and 3’-to-5’ exonuclease activities. Western blotting analysis using anti-BMRF 1 gene product antibody will clarify whether these proteins are BMRF 1 gene product or not. The 3’-to-5’ exonuclease active site of f. co/i DNA polymerase I is predicted to be conserved for both prokaryotic and eukaryotic DNA polymerases based on amino acid sequence homology (24). In fact, it has been reported that HSV DNA polymerase has the 3’-to5’ exonuclease activity, which exhibits proofreading function in the polymerization process like prokaryotic DNA polymerases (25). In the case of Epstein-Barr
381
SHORT COMMUNICATIONS TABLE3
tance. This work was supported by the research fund of lshida Foundation and partly by the research fund of Ryutaro Kato.
dNMP TURNOVERBY EBV DNA POLYMERASE
Enzyme Experiment A EBV DNA polymerase Poll, large fragment Experiment B EBV DNA polymerase Poll, large fragment
dNMP incorporated km-d
dNMP released bmol)
374 582
8 5
32.3 168
2.0 31
dNMP turnover (%I
2.1 0.85 5.8 15.5
Note. For pucleotide turnover analysis, the assays were the same as DNA polymerase assay. The template primer was either 80 fig/ml activated DNA or 80 pglml poly(dC).oligo(dG),,_,,. When poly(dC). oligo(dG) was used as template primer (Experiment A), the reaction mixture contained 40 &I [3H]dGTP as substrate and the reaction was done in the absence of ammonium sulfate. When activated DNA was used (Experiment B), the reaction (25 ~1) contained 40 PM [3H]TTP, 40 pM other dNTPs, and 80 mM ammonium sulfate. After 30 min incubation, the tubes were transferred to ice. Twenty microliters was transferred to tubes containing 5 ~1 of a solution containing 10 mM of dTMP or dGMP as markers and 100 mM EDTA. This mixture was spotted onto polyethyleneimine-cellulose plastic sheet (Merck&Co, Inc.) and developed in 1 M formic acid and 0.5 M LICI (7 7). Position of dTMP or dGMP was cut out and its radioactivity was determined by scintillation counting. The counting efficiency was 37%. The remaining 5 ~1 of the original 25 ~1 of reaction volume was used to determine nucleotide incorporation. Measurements of nucleotide turnover were performed as follows (27): dNMP turnover = (dNMP released) X (dNMP incorporated + dNMP released)-’ x 100.
virus, Grossberger and Clough (9) have reported that EBV DNA polymerase has nucleotide turnover activity caused by contained nuclease activity. But it had been unclear whether the nuclease activity was really polymerase-associated exonuclease or contaminating DNase. This study revealed that the 3’-to-5’ exonuclease is closely associated with EBV DNA polymerase and that the exonuclease activity was different from that of EBV-induced DNase in sensitivity to ammonium sulfate and chromatographic patterns. Moreover, the 3’-to-5’ exonuclease associated with EBV DNA polymerase could excise the terminal mismatched nucleotide rapidly, suggesting that major role of the 3’-to-5’ exonuclease is to monitor whether the correct deoxyribonucleotide has been incorporated during replication. ACKNOWLEDGMENTS I thank Drs. K. Maeno, Y. Nishiyama. and I. R. Lehman for encouragement. I thank T. Tsuruguchi and E. lwata for technical assis-
REFERENCES 1. ZUR HAUSEN, H., In “DNA Tumor Viruses” (J. Tooze, Ed.), pp. 747-795. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1981. YATES. J. L., WARREN, N.. REISMAN, D., and SUGDEN, B., Proc. Narl. Acad. Sci USA 81, 3806-3810 (1984). OOKA, T., TURENNE, M., THE, G., and DAILLIE, J., J. Vifol. 49, 626628 (1984). HAMMERSCHMIDT,W.. and SUGDEN, B., Cell 55,427-433 (1988). BAER, R., BANKIER,A. T., BIGGIN, M. D.. DEININGER,P. L., FARRELL, P. J., GIBSON. T. J., HATFULL, G., HUDSON, G. S., SATCHWELL, S. C., SEGUIN. C., TUFFNELL, P. S., and BARRELL,B. G., Nature 310,207-211(1984). 6. KALLIN, B., STERNAS,L., SAEMUNDSSEN,A. K., LUKA, J., JORNVALL, H., ERIKSSON.B., TAO, P-Z.. NILSSON, M. T., and KLEIN, G., J. Viral. 54, 561-568 (1985). 7. DANA, A. K., FEIGHNY, R. J., and PAGANO, J. S., J. Biol. Chem. 255, 5120-5125 (1980). 8. GOODMAN, S. R., PREZYNA,C., and CLOUGH, W., J. B/o/. Chem. 253,8617-8628(1978). 9. GROSSBERGER,D., and CLOUGH, W., Biochemistry 20, 40494055 (1981). 10. LI. J-S., ZHOU, B-S., DUTSCHMAN, G. E., GRILL, S. P.. TAN, R-S., and CHENG, Y-C.. 1. Viral. 61, 2947-2949 (1987). 11. MILLER, R. L., GLASER, R., and RAPP, F., Virology 76, 494-502 (1977). 12. OOKA, T., LENOIR, G. M., DECAUSSIN.G., BORNKAMM, G. W.. and DAILLIE, J., J. Viral. 58, 671-675 (1986). 13. LI, 1. S., and CHENG, Y. C.. Virus Genes 1, 369-375 (1988). 14. FRY, M., and LOEB. L. A., “Animal Cell DNA Polymerases” CRC Press, Boca Raton, FL, 1986. 15. LEE, M. Y. W. T., TAN, C-K., DOWNEY, K. M., and So, A. G., Biochemistry23, 1906-1913 (1984). 16. NISHIYAMA, Y., MAENO, K., and YOSHIDA, S., I/irology 124, 221231 (1983). 17. O’DONNELL, M. E., ELIAS, P., and LEHMAN, I. R.. J. Biol. Chem. 262,4252-4259 (1984). 18. KNOPF, K-W., EUR. J. B/ocHE~~. 98, 231-244 (1979). 19. CRUTE, J. J., and LEHMAN, I. R., J. Biol. Chem. 264, 19,26619,270 (1989). 20. BRUTLAG, D., and KORNBERG,A., 1. Biol. Chem. 247, 241-248 (1972). 21. KORNEIERG, A., “DNA REPLICATION” FREEMANAND Co., SAN FRANCISCO, 1980. 22. PEARSON, G. R., VROMAN, B., CHASE, B., SCULLEY. T., HUMMEL. M., and KIEFF. E., J. Viral. 47, 193-201 (1983). 23. CHIOU, 1. F., LI, 1. K. K., and CHENG, Y. C., Proc. Nat/. Acad. SC;. USA82,5728-5731(1985). 24. BERNAD, A., BLANCO, L., LAZARO, J. M.. MARTIN, G., and SALAS, M., Ce// 59, 219-228 (1989). 25. ABBOTT? J., NISHIYAMA, Y., YOSHIDA, S., and LOEB, L. A., Nucleic Acids Res. 15, 1 185-l 198 (1987). 26. MCMACKEN. R., UEDA, K., and KORNBERG,A., Proc. Nat/. Acad. SC;. USA 74,4190-4194 (1977). 27. KAGUNI, L. S., DIFRANCESCO, R. A., and LEHMAN, I. R., J. Biol. Chem.259,9314-9319(1984).
182,
376-381
(1991)
Characterization of 3’40-5 with Epstein-Barr
Exonuclease Activity Associated Virus DNA Polymerase
TATSUYA
TSURUMI
Laboratory of Virology, Research Institute for Disease Mechanism and Control, Nagoya University School of Medicine, Showa-ku, Nagoya 466, Japan Received November
6, 1990; accepted January 24, 199 1
Epstein-Barr virus (EBV) DNA polymerase mediates viral DNA replication during the lytic phase of the EB virus life cycle. In order to characterize its enzymatic activities EBV DNA polymerase was purified more than 1200-fold from chemically induced 895-8 cells. One polypeptide with molecular weight of 110,000 corresponded to the predicted EBV DNA polymerase, whereas the other polypeptides did not. A 3’-to-5’ exonuclease activity was copurified with the EBV DNA polymerase through the course of the purification. Unlike HSV DNA polymerase, 5’-to-3 exonuclease activity was not associated with the EBV DNA polymerase on the final step chromatography of single-stranded DNA agarose column. The associated 3’-to-5’ exonuclease activity was stimulated by ammonium sulfate like the polymerase activity. It exhibited DNA-dependent nucleotide turnover activity and preferentially excised a terminal mismatched nucleotide on hybridized polynucleotides compared to the correctly paired substrate, indicating that the 3’-to-5 exonuclease may play a role in proofreading in the polymerization process. 0 1991 Academic Press, Inc.
Epstein-Barr virus (EBV) is a human B lymphotropic herpesvirus which is a causative agent of infectious mononucleosis and is known to be closely associated with African Burkitt’s lymphoma and nasopharyngeal carcinoma (1). EBV infects and immortalizes B lymphocytes, in which only a limited number of viral genes are expressed and the EBV genome is maintained as a circular plasmid molecule by using host DNA polymerases. OriP, an origin of EBV, mediates this type of replication during the latent phase of EBV life cycle (2). In a small fraction of the immortalized B cells most of viral genes are expressed to yield the lytic phase of the life cycle. Chemical agents such as 12-0-tetradecanoyl phorbol-13-acetate (TPA) or sodium n-butyrate can induce the lytic phase itn some cell lines latently infected with EBV (3). During the lytic phase of viral life cycle EBV DNA is amplified 100 to 1000 times. The replication product is a head-to-tail concatemer, which is synthesized by EBV DNA polymerase presumably via rolling circle mechanism initiated from oriLyt (4). Thus, EBV DNA polymerase acts as a key enzyme during the lytic phase of EBV DNA replication. The complete nucleotide sequence of the B95-8 strain of EB virus was determined by Baer et a/. and likely protein-coding regions were identified (5). The genome of EBV B95-8 consists of 172,281 bp and encodes 84 proteins. EBV DNA polymerase gene was identified and localized on the BALF5 open reading frame of the EBV genome (5, 6). Estimated molecular weight from the open reading frame is 1 13,400 Da. 0042-6822/91
$3.00
Copyright 0 1991 by Academic Press. Inc. All rights of reproduction !n any form reserved
EBV-induced DNA polymerase has been partially purified (6- 12) and characterized with respect to reaction optima, sensitivities to various inhibitors, and template specificities. However, the enzymatic characterization of EBV DNA polymerase is not enough to understand the DNA synthesis process, especially with respect to its associated nuclease activity. In this study I have purified EBV DNA polymerase from B95-8 cells treated with both TPA and sodium n-butyrate and characterized EBV DNA polymerase with respect to its associated exonuclease activity in detail. B95-8 cells were treated with 100 rig/ml TPA and 5 mM sodium n-butyrate to induce the lytic phase of EBV DNA replication. Salt-stimulated DNA polymerase activity was induced after addition of the chemical agents. The DNA polymerase was extracted from the nuclei of induced B95-8 cells. The salt-stimulated DNA polymerase activity was detected neither in the cytosol fraction nor in the low-salt extract of the nuclei, but fractionated into 0.4 M NaCl extract of the nuclei. This first step purification was very useful to separate the EBV DNA polymerase from host DNA polymerase (Y, which was fractionated into cytosol, and to increase purification (Table 1). Figure 1 shows the purification of the viral DNA polymerase by sequential chromatography on phosphocellulose (A), heparin agarose (B), and single-stranded DNA agarose (C). On the phosphocellulose column chromatography EBV DNA polymerase eluted at 0.35 n/l NaCI. A DNase eluted at 0.25 M NaCI. The DNase could be EBV-induced DNase because its 376
SHORT COMMUNICATIONS
377
TABLE 1 SUMMARYOFPURIFICATIONOF EBV DNA POLYMERASE
Fraction I II Ill IV V
Step Total cell extract (Cytosol + nuclei) Medium salt nuclear extract Phosphocellulose Heparin agarose ssDNA agarose
Total protein (w)
Total activity (units)
Specific activity (units/mg)
Overall yield (%)
Purification (fold)
238
568,000
2,386
100
1
20
492,800
24,640
87
19
1.37 0.494 0.052
414,720 268,800 157,500
302,715 544,130 3,028,846
73 47 28
127 228 1269
Note. One unit is defined as 1 pmol of dNMP incorporation
for 60 min.
activity was 8 times as high as the activity observed in case of uninduced 695-8 cell extracts. On heparin agarose column chromatography EBV DNA polymerase eluted at 0.42 M NaCl and was completely separated from the contaminating DNase activity, which eluted at 0.35 M NaCI. However, some nuclease activity was detected along with the polymerase peak (Fig. 1B). On the single-stranded DNA agarose column, the DNA polymerase eluted at 0.35 M NaCI. The summary of the purification of EBV DNA polymerase was given in Table 1. The EBV DNA polymerase was purified 1269-fold from the whole cell extract with a yield of 28%. The peak fraction of the EBV DNA polymerase on the single-stranded DNA agarose column chromatography contained three protein bands, of which a 110. kDa polypeptide coincided with the peak of the polymerase activity (Fig. 1 D). The remaining two proteins (50 and 48 kDa) eluted with peaks at 0.25 M NaCl on the single-stranded DNA agarose column, of which fractions were lacking the DNA polymerase activity. Thus, only the 1 10-kDa protein corresponded to the predicted EBV DNA polymerase. The peak fractions of the EBV DNA polymerase were devoid of detectable primase and ATPase activities. The enzyme exhibited a broad optimum activity between pHs 6.8 and 9.2. The DNA polymerase activity was stimulated more than 2-fold by 75 to 100 mM ammonium sulfate and was resistant even at 150 mM ammonium sulfate, while DNA polymerase (Ywas very sensitive to ammonium sulfate (50% inhibition at 30 mM). The EBV DNA polymerase utilized poly(dC). oligo(dG) 29-fold more efficiently than activated DNA, but the RNA template-DNA primer, poly(rA) . oligo(dT), could not be utilized at all. Further, the polymerase activity was inhibited by 70% at 10 pg/ml phosphonoacetic acid and by 80% at 0.5 mM A/-ethylmaleimide. Aphidicolin was also inhibitory to the EBV DNA polymerase as well as to DNA polymerase CL 50% inhibition was observed at 2 pg/ml. The above enzymatic
properties were specific for EBV DNA polymerase as was reported previously (6, 12, 13) and clearly distinguishes the viral enzyme from the host DNA polymerase LY,6, y, or 6 (6, 14, 75). Recent studies have shown that 3’-to-5’exonuclease activity is associated with other herpesvirus DNA polymerases ( 16- 18). To analyze the nuclease activity associated with the EBV DNA polymerase in detail, 3’-to5’ exonuclease activity on fractions of the phosphocellulose, heparin agarose, and single-stranded DNA agarose column chromatography were assayed using 3’-terminal labeled poly(dA-dT) as substrate in the presence of 80 mM ammonium sulfate. As shown in Fig. 1, the 3’-to-5’ exonuclease activity was found to cochromatograph with the EBV DNA polymerase through the course of the purification. Recently, it has been reported that herpes simplex virus type 1 DNA polymerase contains not only 3’-to-5’ exonuclease but also 5’-to-3’ exonuclease (19). To determine this possibility for the EBV DNA polymerase, 5’-to-3’ exonuclease (Fig. 1C) was assayed by using [5’-32P]poly(dA),,, . poly(dT),ooo as substrate. However, 5’-to-3’exonuclease eluted ahead of the EBV DNA polymerase and was not associated with the EBV DNA polymerase fractions on the final step of chromatography on single-stranded DNA agarose column. The 3’-to-5’ exonuclease activity was assayed by using 3’-terminally labeled poly(dA-dT) alternating copolymer as substrate. The reaction product was dTMP determined by PEI thin-layer chromatography (data not shown), indicating that the EBV DNA polymerase hydrolyzed 3’-terminal nucleotide. The effect of adding increasing amounts of ammonium sulfate was determined (Fig. 2). The polymerase associated 3’-to-5’exonuclease was stimulated by 25 to 100 mM ammonium sulfate. Even at 150 mM ammonium sulfate 80% of the activity remained. This pattern of stimulation by ammonium sulfate was analogous to that observed with the EBV DNA polymerase activity. On the other hand, the
SHORT COMMUNICATIONS
378
(u!w OZ/~.O t x loud) ~NI=
aseJawA@
t/NO MEI
l
(c-or x WW I+ aiqnlos p!zv :aseNa
,-yy=-=j-jB
o
SHORT COMMUNICATIONS
nuclease activity of EBV-induced DNase, which eluted ahead of EBV DNA polymerase on the phosphocellulose column chromatography in the same assay system, was more sensitive to the salt (Fig. 2). Fifty percent inhibition was observed at 60 mA4 ammonium sulfate. Thus, the 3’-to-5’ exonuclease activity associated with the viral DNA polymerase was totally different from that of EBV-induced DNase. The other properties of the 3’-to-5’ exonuclease are summarized in Table 2. Mg2+ was essential for the reaction and could be replaced by 1 mM Mn2+. The exonuclease activity was very sensitive to 3 mn/l NEM. However, aphidicolin and phosphonoacetic acid had almost no inhibitory effect on the exonuclease activity at a concentration of 15 pg/ml and 13 pg/ml, respectively, at which EBV DNA polymerase activitywas inhibited by 809/o. Brutlag and Kornberg (20) demonstrated the proofreading activity of E. co/i polymerase I by measuring the removal of a terminal mismatched nucleotide. To determine whether the 3’-to-5 exonuclease associated with EBV DNA polymerase has proofreading activity, removal of 3’-terminal labeled nucleotide from
pob(dA)~(dT)20[3Hl(dT),., or wM@J - (dT)2,[3WG),.2 was examined (Fig. 3). The paired 3’ terminus was stable to the 3’-to-5’ exonuclease in the presence of dlTP, but excision of the correctly paired terminus in
379
the absence of dlTP was observed (Fig. 3B). This is most likely due to the addition of nucleotide to the paired 3’terminus by the polymerase. In contrast, even in the presence of dTTP the 3’-to-5’ exonuclease rapidly removed the unpaired 3’-terminal nucleotide (Fig. 3A). The 3’-to-5’ exonuclease hydrolyzed an unpaired 3’terminal nucleotide at about twice the rate at which the paired 3’-terminal nucleotide was hydrolyzed. These observations suggest that excision of the mismatched terminal nucleotide is necessary before elongation can proceed. DNA-dependent nucleotide turnover is a characteristic of DNA polymerases possessing 3’-to-5’exonucleases (21). EBV DNA polymerase exhibited nucleotide turnover from deoxynucleotide triphosphate to monophosphate form when poly(dC) . oligo(dG) or activated DNA was used as template primer (Table 3). Both turnover assays were done in the presence of controls, which were identical assays performed in the absence of the templates. No deoxynucleotide monophosphate was produced in such control experiments (data not shown), demonstrating that this reaction was template dependent. The dGMP or dTMP is presumably formed upon hydrolysis of paired 3’termini during DNA synthesis as observed previously for the HSV DNA polymerase (18) and for other DNA polymerases possessing 3’90-5 exonuclease activities (2 I).
FIG. 1. Purification of EBV DNA polymerase. The 895-8 cells were seeded at 1O6 cells/ml in culture medium. After 30 hr of cultivation, the chemical inducers were added at final concentrations of 100 rig/ml for TPA and 5 mM for sodium n-butyrate. The cells were harvested at 48 hr postinduction, washed, and then suspended in hypotonic lysis buffer (20 mM Na-HEPES, pH 7.2, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF, 10 pg/ml pepstatin A, and 10 pg/ml leupeptin) for 30 min at 0”. The nuclei were prepared by Dounce homogenization and finally suspended in the stock buffer (20 mM Na-HEPES, pH 7.2, 1 mM EDTA, 1 mM EGTA. 1 mM DTT, 1 mM PMSF, 10 pg/ml pepstatin A, 10 pglml leupeptin, and 10% sucrose) and stored at -80”. The stored nuclear preparations were thawed and collected by low-speed centrifugation. The pellet was suspended in buffer A (20 mM Tris-HCI, pH 7.2, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF, 10 pg/ml pepstatin A, 10 pg/ml leupeptin) containing 0.4 M NaCl and centrifuged at 20,000 rpm in a Hitachi SW28 rotor for 1 hr at 4”. The supernatant was dialyzed against buffer A containing 0.1 M NaCl overnight (Fraction II). (A) Fraction II was loaded onto a phosphocellulose (P 1 1, Whatman BioSystems Ltd.) column (5 ml bed volume) equilibrated with buffer A containing 0.1 M NaCI. The phosphocellulose column was washed with 25 ml buffer A containing 0.1 M NaCl and eluted with a 60.ml linear gradient from 0.1 to 0.8 M NaCl In bufferA. Peak fractions were pooled and dialyzed against 100 vol of buffer A containing 0.1 M NaCl (Fraction Ill). (B) Fraction Ill was applied to a column (1.5 ml bed volume) of heparin agarose equilibrated with buffer A containing 0.1 M NaCI. The column was washed and eluted with a 30.ml linear gradient from 0.1 to 1 .O M NaCl in bufferA. Active fractions were pooled and dialyzed against 100 vol of buffer A containing 0.1 M NaCl (Fraction IV). (C) Fraction IV was applied to a column (1 ml bed volume) of single-stranded DNA agarose equilibrated with bufferA containing 0.1 M NaCI. The column was washed and eluted with a 24-ml linear gradient from 0.1 M to 0.8 M. EBV DNA polymerase (0) was assayed in reaction mixture (25 pl) that contained 50 mM Tris-HCI, pH 8.4, 80 mM ammonium sulfate, 10% glycerol, 6 mM MgCI,, 100 pg/ml bovine serum albumin, 80 pg/ml activated calf thymus DNA, 1 mM DTT, 25 PM each of dATP, dGTP, and dCTP, and 10 FM [3H]lTP (1800 cpm/pmol). The reaction was started by the addition of 2 ~1of enzyme fraction, After 20 mm at 35”, TCA precrpitable counts were determined. Assay for DNase activity (0) was done with E. co/i DNA labeled with [3H]thymidine to a specific actrvity of 38,000 cpm/pg. Each assay of 25 ~1 contained 1 pg of labeled DNA, 50 m/L?Tris-HCI, pH 8.4, 10% glycerol, 6 mM MgCI,, 100 rg/ml BSA, 1 mM DTT. The reaction was started by the addition of 4 ~1 of enzyme fraction, After 30 min at 35”, released nucleotides were determined (19). 3’-to-5’ exonuclease activity (0) was assayed as follows, Reaction mixture (25 ~1) contained 40 mM Tris-HCI, pH 8.4, 80 m/L? ammonium sulfate, 6 mM MgCI,, 100 pglml BSA, 1 mM DTT, and 2 pg/ml [3’-3H]poly(dApdT) (1.5 X 1O5 cpmlpg). Four microliters of enzyme fraction was added to start the reaction. After 10 min, released nucleotides were determined (76). 5’.to-3’ exonuclease activity (A) was assayed in reaction mixture (25 ~1) containing 40 mM Tris-HCI. pH 8.4, 3 mM MgCI,, 100 @g/ml BSA, 1 mM DlT, and 10 pM [5’-32P]poly(dA)~poly(dT),,,, (445 cpmipmol terminal nucleotide). Four microliters of each fraction was added to start the reaction. After 20 mm, released nucleotides was determined (19). (D) Silver staining of SDS-PAGE of EBV DNA polymerase. Aliquots of the peak fraction (26) of the srngle-stranded DNA agarose column subjected to 8% SDS-polyacrylamide gel electrophoresis and stained with srlver. DF IS the dye front. The positions of the molecular weight standards (M, X 1Om3)are indicated by arrows.
380
SHORT COMMUNICATIONS B
--e-.-. -T-
\
T’
-dd-di-dA-
0 \
OlO
I--
t 0-
20
Time ( min )
FIG. 3. The 3’.to-5’ exonuclease activity of EBV DNA polymerase on terminally labeled nucleotides. The 3’-to-5’ exonuclease activity on the 3’.unpaired (A) and -paired (B) substrates were assayed in the presence (0) or absence (0) of dTTP. The reaction contained
0 0
50
100
(NH+S04
150
(mM)
FIG. 2. Effect of ammonium sulfate on the 3’.to-5’ exonuclease activity associated with EBV DNA polymerase and that of EBV-induced DNase. The r-to-5 exonuclease activity of EBV DNA polymerase (0) and that of EBV-induced DNase (0) were assayed using [3’-3H]poly(dA-dT) as substrate as described in the legend to Fig. 1 except that ammonium sulfate concentration was varied as indicated. The peak fractions of the DNase which eluted ahead of EBV DNA polymerase on the phosphocellulose column chromatograph (Fig. 1A) were applied to heparin agarose column chromatography. The peak fraction of the DNase on the heparin agarose column chromatograph was used as the sample for the EBV-induced DNase. Two microliters of enzyme fraction was applied to start the reaction. The activity is expressed as percentage of the activity obtained in the absence of ammonium sulfate.
The peak fraction of EBV DNA polymerase on the single-stranded DNA agarose column chromatograph contained three protein bands, of which the 1 lo-kDa polypeptide corresponded to the polymerase activity and the 3’-to-5’ exonuclease activity. It was in agreement with estimated size from the open reading frame
TABLE 2 CHARACTERISTICSOF THE 3’-TO-~’ EXONUCLEASEACTIVITV ASSOCIATED WITH EBV DNA POLYMERASE
Omission or addition None Omit Omit Omit Add Add
(complete) Mgz+ Mg*+ plus Mn2+ (1 mM) DTr plus N-ethylmaleimide (3 mM) aphidicolin (15 pglml) phosphonoacetic acid (13 pglml)
40
Time ( min )
dTMP released (% of maximum) 100 2 98 4 88 89
Note. The 3’-to-5’ exonuclease activity of the peak fraction (26) of EBV DNA polymerase on the single-stranded DNA agarose column chromatography was measured using [Y-3H]poly(dA-dT) as described in the legend to Fig. 1.
poWA) *kWd3WdG), 2 (4 or poWA) *W)20-[3H1(dT)1.4 (W as substrate. Excision of a terminal mismatch was examined as described by Brutlag and Kornberg (79). Reaction mixture (25 pl) contained 40 mM Tris-HCI, pH 8.4, 1 mM DTT, 6 mM MgCI,, 100 pg/ml BSA, 80 mM ammonium sulfate, and 0.1 pg of template primer, which was either poly(dA) . (dT)2,-[3H]dT,., (704 cpm/pmol terminal nucleotide) or poly(dA). dT,JH]dG,,, (1059 cpm/pmol terminal nucleotide). When present, dTTP was 100 pM each. Reactions were carried out at 30” to reduce partial denaturation of the primer template. The exonuclease activity was measured by the reduction in DE 81 absorbed radioactivity (26).
of BALF5. It has recently found that herpes simplex virus type 1 DNA polymerase is a heterodimer complex which consists of the products of UL30 (Pol) and UL42 (Pol accessory protein) genes (19). In the case of Epstein-Barr virus, it has been reported that BMRF 1 gene product has an ability to stimulate the EBV DNA polymerase activity (10, 22, 23). Moreover, BMRF 1 gene product possesses some amino acid sequence homology with HSV-1 UL42 gene product (T. Tsurumi, unpublished data). The estimated size of BMRF 1 gene product is 43K (5). The peak fraction of EBV DNA polymerase on the single-stranded DNA agarose column chromatography contained two other polypeptides, 50 and 48 kDa, which did not correlate with the polymerase and 3’-to-5’ exonuclease activities. Western blotting analysis using anti-BMRF 1 gene product antibody will clarify whether these proteins are BMRF 1 gene product or not. The 3’-to-5’ exonuclease active site of f. co/i DNA polymerase I is predicted to be conserved for both prokaryotic and eukaryotic DNA polymerases based on amino acid sequence homology (24). In fact, it has been reported that HSV DNA polymerase has the 3’-to5’ exonuclease activity, which exhibits proofreading function in the polymerization process like prokaryotic DNA polymerases (25). In the case of Epstein-Barr
381
SHORT COMMUNICATIONS TABLE3
tance. This work was supported by the research fund of lshida Foundation and partly by the research fund of Ryutaro Kato.
dNMP TURNOVERBY EBV DNA POLYMERASE
Enzyme Experiment A EBV DNA polymerase Poll, large fragment Experiment B EBV DNA polymerase Poll, large fragment
dNMP incorporated km-d
dNMP released bmol)
374 582
8 5
32.3 168
2.0 31
dNMP turnover (%I
2.1 0.85 5.8 15.5
Note. For pucleotide turnover analysis, the assays were the same as DNA polymerase assay. The template primer was either 80 fig/ml activated DNA or 80 pglml poly(dC).oligo(dG),,_,,. When poly(dC). oligo(dG) was used as template primer (Experiment A), the reaction mixture contained 40 &I [3H]dGTP as substrate and the reaction was done in the absence of ammonium sulfate. When activated DNA was used (Experiment B), the reaction (25 ~1) contained 40 PM [3H]TTP, 40 pM other dNTPs, and 80 mM ammonium sulfate. After 30 min incubation, the tubes were transferred to ice. Twenty microliters was transferred to tubes containing 5 ~1 of a solution containing 10 mM of dTMP or dGMP as markers and 100 mM EDTA. This mixture was spotted onto polyethyleneimine-cellulose plastic sheet (Merck&Co, Inc.) and developed in 1 M formic acid and 0.5 M LICI (7 7). Position of dTMP or dGMP was cut out and its radioactivity was determined by scintillation counting. The counting efficiency was 37%. The remaining 5 ~1 of the original 25 ~1 of reaction volume was used to determine nucleotide incorporation. Measurements of nucleotide turnover were performed as follows (27): dNMP turnover = (dNMP released) X (dNMP incorporated + dNMP released)-’ x 100.
virus, Grossberger and Clough (9) have reported that EBV DNA polymerase has nucleotide turnover activity caused by contained nuclease activity. But it had been unclear whether the nuclease activity was really polymerase-associated exonuclease or contaminating DNase. This study revealed that the 3’-to-5’ exonuclease is closely associated with EBV DNA polymerase and that the exonuclease activity was different from that of EBV-induced DNase in sensitivity to ammonium sulfate and chromatographic patterns. Moreover, the 3’-to-5’ exonuclease associated with EBV DNA polymerase could excise the terminal mismatched nucleotide rapidly, suggesting that major role of the 3’-to-5’ exonuclease is to monitor whether the correct deoxyribonucleotide has been incorporated during replication. ACKNOWLEDGMENTS I thank Drs. K. Maeno, Y. Nishiyama. and I. R. Lehman for encouragement. I thank T. Tsuruguchi and E. lwata for technical assis-
REFERENCES 1. ZUR HAUSEN, H., In “DNA Tumor Viruses” (J. Tooze, Ed.), pp. 747-795. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1981. YATES. J. L., WARREN, N.. REISMAN, D., and SUGDEN, B., Proc. Narl. Acad. Sci USA 81, 3806-3810 (1984). OOKA, T., TURENNE, M., THE, G., and DAILLIE, J., J. Vifol. 49, 626628 (1984). HAMMERSCHMIDT,W.. and SUGDEN, B., Cell 55,427-433 (1988). BAER, R., BANKIER,A. T., BIGGIN, M. D.. DEININGER,P. L., FARRELL, P. J., GIBSON. T. J., HATFULL, G., HUDSON, G. S., SATCHWELL, S. C., SEGUIN. C., TUFFNELL, P. S., and BARRELL,B. G., Nature 310,207-211(1984). 6. KALLIN, B., STERNAS,L., SAEMUNDSSEN,A. K., LUKA, J., JORNVALL, H., ERIKSSON.B., TAO, P-Z.. NILSSON, M. T., and KLEIN, G., J. Viral. 54, 561-568 (1985). 7. DANA, A. K., FEIGHNY, R. J., and PAGANO, J. S., J. Biol. Chem. 255, 5120-5125 (1980). 8. GOODMAN, S. R., PREZYNA,C., and CLOUGH, W., J. B/o/. Chem. 253,8617-8628(1978). 9. GROSSBERGER,D., and CLOUGH, W., Biochemistry 20, 40494055 (1981). 10. LI. J-S., ZHOU, B-S., DUTSCHMAN, G. E., GRILL, S. P.. TAN, R-S., and CHENG, Y-C.. 1. Viral. 61, 2947-2949 (1987). 11. MILLER, R. L., GLASER, R., and RAPP, F., Virology 76, 494-502 (1977). 12. OOKA, T., LENOIR, G. M., DECAUSSIN.G., BORNKAMM, G. W.. and DAILLIE, J., J. Viral. 58, 671-675 (1986). 13. LI, 1. S., and CHENG, Y. C.. Virus Genes 1, 369-375 (1988). 14. FRY, M., and LOEB. L. A., “Animal Cell DNA Polymerases” CRC Press, Boca Raton, FL, 1986. 15. LEE, M. Y. W. T., TAN, C-K., DOWNEY, K. M., and So, A. G., Biochemistry23, 1906-1913 (1984). 16. NISHIYAMA, Y., MAENO, K., and YOSHIDA, S., I/irology 124, 221231 (1983). 17. O’DONNELL, M. E., ELIAS, P., and LEHMAN, I. R.. J. Biol. Chem. 262,4252-4259 (1984). 18. KNOPF, K-W., EUR. J. B/ocHE~~. 98, 231-244 (1979). 19. CRUTE, J. J., and LEHMAN, I. R., J. Biol. Chem. 264, 19,26619,270 (1989). 20. BRUTLAG, D., and KORNBERG,A., 1. Biol. Chem. 247, 241-248 (1972). 21. KORNEIERG, A., “DNA REPLICATION” FREEMANAND Co., SAN FRANCISCO, 1980. 22. PEARSON, G. R., VROMAN, B., CHASE, B., SCULLEY. T., HUMMEL. M., and KIEFF. E., J. Viral. 47, 193-201 (1983). 23. CHIOU, 1. F., LI, 1. K. K., and CHENG, Y. C., Proc. Nat/. Acad. SC;. USA82,5728-5731(1985). 24. BERNAD, A., BLANCO, L., LAZARO, J. M.. MARTIN, G., and SALAS, M., Ce// 59, 219-228 (1989). 25. ABBOTT? J., NISHIYAMA, Y., YOSHIDA, S., and LOEB, L. A., Nucleic Acids Res. 15, 1 185-l 198 (1987). 26. MCMACKEN. R., UEDA, K., and KORNBERG,A., Proc. Nat/. Acad. SC;. USA 74,4190-4194 (1977). 27. KAGUNI, L. S., DIFRANCESCO, R. A., and LEHMAN, I. R., J. Biol. Chem.259,9314-9319(1984).
