THE JOURNAL OF Bmm~~car. CHEMISTRY Vol. 247,No. 16, Issue of August25, pp. 50%5033,1972

Printed in U.S.A.

Purification and Properties of Nuclear and Cytoplasmic

Deoxyribonucleic Acid Polymerases from

Human KB Cells*

(Received for publication, March 28, 1972)

W. DAVID SEDWICK,~ TERESA SHU-FONG WANG, AND DAVID KORN

From the Department of Pathology, Stanford University School of Medicine, Stanford, California 94SO5

SUMMARY

We describe the properties of three DNA polymerases isolated from nuclear and cytoplasmic fractions of freshly harvested KB cells. Two of the enzymes, cytoplasmic poly- lymerase and nuclear polymerase Nl, have been highly purified, one of them to apparent homogeneity. The third enzyme, nuclear polymerase N2, which is only partially purified, resembles the cytoplasmic enzyme in its physical properties but possesses distinctive enzymatic characteristics that suggest it may represent an independent third protein. The KB enzymes initiate polymerization at 3’-hydroxyl termini and resemble Escherichia coli DNA polymerase II in their obligatory requirement for gap-containing templates. None of the enzymes can copy the ribonucleotide strand of hybrid templates, even under specific conditions in which E. coli DNA polymerase I does demonstrate such activity. They can, however, with low efficiency copy a deoxynucleotide strand using an oligoribonucleotide as primer. Elucidation of the possible in vivo roles of these enzymes in DNA replica- tion, recombination, and repair must await further bio- chemical and genetic studies.

At present our understanding of the enzymatic basis of DNA replication in prokaryotic or eukaryotic cells is incomplete. Pre- vious concepts, centered on the role of Escherichia coli DNA polymerase I, have been altered by the discovery of the polA mutation (I), by failure to detect changes in polymerase I ac- tivity in various DNAt” mutants at restrictive temperatures, and by the recognition that cells of E. coli contain at least two additional DNA polymerases, II and III (2-5). Although evi- dence has been presented that implicates each of the latter two enzymes in in vivo DNA replication (6, 7), the relationships of the t,hree recognized E. coli DNA polymerases to the putative DNA replicase remain to be established.

Since the initial description of a DNA polymerase activity in

* These studies were supported by grants from the American Cancer Society (P-579), the Jane Coffin Childs Memorial Fund for Medical Research (266), and National Institutes of Health (AI- 08806).

$ W.D.S. is a Fellow of the Jane Coffin Childs Fund.

calf thymus (8, 9), studies in eukaryotic cells, influenced by prokaryote models, have been directed toward the isolation of ostensibly single, although only partially purified enzymes that have been characterized in terms of the several well defined properties of homogeneous E. coli polymerase I (10). In recent years, however, the existence of multiple DNA polymerase ac- tivities has been recognized in eukaryotic cells. These include an activity that appears to be associated with mitochondria (11, 12) and, variously, two or three separable activities in nuclei and cytoplasm (13-16). In general, and particularly with re- spect to human cells, none of these activities has been highly purified and its properties rigorously established.

We have previously described (17) a partially purified DNA polymerase from human KB cells, several of whose properties were significantly different from those either of E. co& polym- erase I or of the calf thymus enzyme. We now report the iso- lation and characterization of three distinct DNA polymerases from KB cells, two of which appear to reside in the nucleus and one in the cytoplasm. Two of these enzymes have been highly purified, one of them to apparent homogeneity.

EXPERIMENTAL PROCEDURE

Materials

Unlabeled deoxyribonucleotides were obtained from P-L Bio- chemicals; tritium-labeled deoxyribonucleotides and ribonu- cleotides from Schwartz; salmon sperm DNA from Calbiochem; pancreatic DNase I and micrococcal nuclease from Worthington; poly[d(A-T)] and poly(A) from Miles Laboratories; DEAE- cellulose (DE-52) and phosphocellulose (Pll) from Whatman; Sephadex G-200 from Pharmacia; hydroxylapatite from BioRad; and ampholine carrier ampholytes from LKB. E. coli exonu- clease III (Sephadex G-100 fraction (18)) was provided by I. R. Lehman; E. coli DNA polymerase I (Sephadex G-100 fraction 7 (18)) and poly[d(T)am] by D. Brutlag; poly[d(T)sooo] by P. Modrich, Stanford Medical School; oligo[d(T)l& by S. Aaron- son, N.I.H. and SV40 supercoiled DNA by D. A. Clayton, Stan- ford Medical School.

Methods

Growth of KB Cells

The cell line used in these studies was obtained from Dr. Mau- rice Green, St. Louis University. Cells were grown in suspen-

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5027

sion culture at 37” in h/Zinimal Essential Medium (Auto-pow tions were carried out for 15 min at 37” and then chilled. The medium for suspension culture, Flow Labs), containing 10% reaction mixtures were immediately spread for electron micro- heat-inactivated calf serum (Gibco). Cultures were routinely scopic examination and grids prepared as previously described seeded at 1.5 to 2.0 x 10” cells per ml and harvested in log phase (23). About 150 DNA molecules were scored in each assay. at 3 to 6 x IO5 cells per ml. The yield of cells was about 1 g Persistence of the initial percentage of Form I molecules indi- wet weight per 2 x lo8 cells. The cultures were periodically cates the absence of endonuclease activity. monitored and found to be free of mycoplasma contamination by L. Hay-flick, Stanford Medical School. Preparation of Column Materials

Nuclease Treatment of Primer-template DNA

Activated salmon sperm DNA was prepared with pancreatic DNase I (19) under conditions that rendered 20% of the DNA acid soluble. Exonuclease III digestion of salmon sperm DNA was performed as described by Kornberg and Gefter (4), and micrococcal nuclease digestion, as described in the Worthington Manual.

Assay of KB Cell DNA Pobymerases

The standard assay for each of the three enzymes contained the following in a total volume of 0.25 ml.

1. Cytoplasmic Enzyme-Tris-HCl, pH 9.2, 10 mM; p-mer- captoethanol, 2.0 mM; bovine serum albumin, 200 pg per ml; MgCl?, 10 mM; activated salmon sperm DNA, 800 pg per ml; dATP, dGTP, dCTP, and dTTP, 50 PM each; [3H]dTTP or [3H]dATP at a final specific activity of 0.04 Ci per mmole; and enzyme.

2. ,Vuclear Polymerase Nd-As above, except unlabeled de- oxgnucleoside triphosphates, 25 pM each, and activated salmon sperm DNA, 100 pg per ml.

S. Nuclear Polymerase NI-As in 1, except MgC12, 20 mM; unlabeled deoxynucleoside triphosphates, 100 pM each; and activated salmon sperm DNA, 100 pg per ml.

The standard assay was carried out for 10 min at 37”. The reaction was terminated by adding 1 ml of water containing 500 pg of salmon sperm DNA and 0.1 M Na pyrophosphate, followed by heating for 2 min at loo”, cooling, and precipitating the DNA with 1 ml of cold 10% trichloroacetic acid. The pre- cipitate was collected on Whatman GF/C glass fiber filters and washed successively with 5% trichloroacetic acid and absolute ethanol. Radioactivity was measured by liquid scintillation spectrometry using a toluene mixture containing 0.5% 5,5- diphenyloxazole and 0.003y0 1,4-bis[2-(5-phenyloxazolyl)]ben- zene.

One unit of enzyme activity is defined as the amount catalyz- ing the incorporation of 1 nmole of labeled deoxynucleotide in 60 min under standard conditions. Specific activity is expressed as units per mg of protein.

Assay of KB Polymerase Fractions for Nuclease Activity

DNA exonuclease activity was measured as previously de- scribed (20) under standard polymerization conditions but in the absence of deoxynucleoside triphosphates. DNA endo- nuclease activity was assayed by a method (21) that involves the determination by electron microscopy of the conversion of covalently closed circular molecules (Form I) of SV40 DNA to open circles (Form II). This conversion is effected by a single endonucleolytic scission of a DNA strand (22). The assays were performed in a total volume of 0.125 ml under standard polymerization conditions but in the absence of DNA primer- template and deoxynucleoside triphosphates. Each reaction contained 0.8 pg of SV40 DNA substrate (consisting of 88% Form I molecules and 12’% Form II molecules) and 1.5 to 1.7 units of the most purified KB polymerase fractions. Incuba-

DEAE (DE-52) was equilibrated by suspension three times in 20 volumes of 0.025 M or 0.05 M potassium phosphate, pH 8.5. Thirty grams (wet weight) of equilibrated DE-52 were stirred with 1 g of bovine serum albumin in 10 volumes of one of the above phosphate buffers for 1 hour at 4”. The slurry was poured into a column and washed with 1 M KzHPO~ until all bovine serum albumin was removed. The DEAE was then washed with water and re-equilibrated as described above.

Phosphocellulose (Pll)-PI1 was prepared as described in the Whatman Manual. The Pll was equilibrated in 0.05 M po- tassium phosphate, pH 7.2 and adsorbed with bovine serum albumin as described above for DE-52, except that 1 M KHzPOd was used to remove the protein. The PI1 was then equilibrated with 0.15 M potassium phosphate, pH 7.2, containing 20% eth- ylene glycol.

Hydroxylapatite-Hydroxylapatite was equilibrated with 0.05 M potassium phosphate, pH 7.2, adsorbed with bovine serum albumin as described above, and then re-equilibrated in 0.05 M

potassium phosphate, pH 7.2. Xephadex G-ZOO-A column (2.5 X 28 cm) was prepared as

described in the Pharmacia Gel Filtration Manual, and equil- ibrated with 0.1 M potassium phosphate, pH 8.5, containing 20% ethylene glycol. The Sephadex G-200 column was washed with 1 g of bovine serum albumin per 20 ml of column bed volume.

Other Methods

Denatured DNA was prepared by heating DNA in 0.02 M

NaCl for 10 mm at 100” and quenching in an ice bath. Protein was determined by a modification of the Lowry method (24) ; spectrophotometrically (25) ; or, for the most purified fraction of nuclear polymerase I, by estimation of Coomassie brilliant blue (R-250) stain intensity in SDS’ acrylamide gels. Vertical slab SDS polyacrylamide gel electrophoresis was carried out according to published methods (26, 27). We are indebted to Dr. P. Greene for introducing us to this technique.

RESULTS

PuriJication of Enzymes

All operations were carried out at O-4”. All buffers contained 1 X low3 M EDTA and 1 x lo+ M P-mercaptoethanol, and those used in procedures after DEAE-chromatography contained in addition 20% ethylene glycol. Enzyme fractions from column steps were routinely concentrated IO-fold by dialysis against solid sucrose at 4”.

Preparation of KB Cell Extracts-Freshly harvested KB cells were suspended for 10 min in 9 volumes of hypotonic solution containing 1 x 10m3 M each of EDTA and P-mercaptoethanol and 2 x lop3 M MgClz and were then broken by gentle shear in a tight fitting glass Dounce homogenizer with the B pestle.

1 The abbreviations used are: SDS, sodium dodecyl sulfate; polymerase C, KB cell cytoplasmic DNA polymerase; polymerase Nl, KB cell nuclear DNA polymerase 1; polymerase N2, KB cell nuclear DNA polymerase 2; dNTP (s), deoxynucleoside triphos- phate (s)

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5028

Breakage was >98y0 complete as monitored by phase micros- copy. The extract was centrifuged at 1,200 g for 10 min to remove nuclei and cell debris, and then at 10,000 g for 15 min to sediment organelles. The supernatant was used as crude extract for purification of the cytoplasmic enzyme.

Polymerase C

Acid Precipitation-Crude extracts were dialyzed against 20 volumes of 0.025 M potassium phosphate, pH 6.0, for 6 hours, and the precipitate that formed in the dialysis bag was discarded. The supernatant was further dialyzed against 20 volumes of 0.2 M sodium acetate, pH 5.5, for 4 hours. The resulting pre- cipitate was collected by centrifugation at 10,000 g for 10 min and solubilized by homogenization into 0.2 RI potassium phos- phate, pH 8.5.

ClirarentriSugation-The solubilized pH 5.5 precipitate was adjusted to 20% (w/v) sucrose, overlayered with 5% (w/v) sucrose in the same buffer and centrifuged in the SW 40.1 rotor for 90 min at 40K rpm at 2”. The fatty pellicles at the top of the 5y0 sucrose layer were removed, and the supernatant solu- tion was collected. The pellet was resuspended in 20% sucrose, 0.2 M potassium phosphate, pH 8.5, by homogenization, and the centrifugation procedure repeated. The two supeinatants were combined.

DEAE-cellulose Column Chromatography-Enzyme fractions were either diluted with water containing 1 x 1OW BI each of EDTX and P-mercaptoethanol to a final ionic strength of 0.05 M potassium phosphate or dialyzed against 0.05 M potassium phosphate, pH 8.5, and added to a DEAE-column (2.5 x 12 cm). The column was washed with 0.05 M potassium phosphate, pH 8.5, until all unadsorbed protein was removed, and then with a 150.ml linear gradient of 0.05 M to 0.2 RI potassium phosphate, pH 8.5. The enzyme activity was eluted at 0.15 M. The active fractions were pooled and concentrated.

Sephadex G-WOO Gel Filtration-The concentrated enzyme fractions were loaded on a Sephadex G-200 column (2.5 x 28 cm)

(V, = 45 ml) in 0.15 M potassium phosphate, pH 7.2, containing 20y0 ethylene glycol. Elution of the column was carried out with the same buffer at a flow rate of 3.6 ml per cm2 per hour. Polymerase C activity was eluted in fractions in the molecular weight range of 110,000 to 140,000, as determined by comparison with marker proteins. Peak fractions of activity were pooled and concentrated.

Phosphocelluloee Column Chromatography-The concentrated enzyme fractions were loaded on a phosphocellulose column (2.5 x 7.5 cm), and the column was washed with 0.15 M potas- sium phosphate, pH 7.2, until all unadsorbed prot,ein was re- moved. Enzyme activity was eluted either wit,h a 7 50.ml lillear gradient of 0.15 RI to 0.40 M of the same buffer or in :L single step with potassium phosphate at 0.21 M. The peak fractions of enzyme activity were pooled and concentmted. They could be stored for more than 6 months over liquid nit,rogen without appreciable loss of activity.

A representative purification of polymemse C is shown in Table I, Part A.

Nuclear Polymeraaes

Purification of Nuclei-The crude nuclear pellet was washed with 9 volumes of buffer containing 0.3% Trit.on X-100, and 1 x 1OW M each of MgC12, fl-mercaptoethanol and potassium phosphate, pH 6.5, in a Dounce homogenizer by 10 strokes with the B pestle. The nuclei were centrifuged at, 1200 g for 10 min and the supernatant discarded. This procedure was repeated once. The resulting purified nuclear pellet wa.s suspended in 9 volumes of sucrose buffer (0.2 nf sucrose, 1 X lo-” M each of MgClz and /Lmercaptoethanol), homogenized as described above, collected by centrifugation at 1200 g for IO rnin and stored at -70”.

Preparation of Nuclear Crude Extract-The purified frozen nuclei were thawed and suspended in 9 volumes of sucrose buffer to which were successively added EDTA, to I x 10e3 11, and solid NaCI, to 4 M. The suspension was vigorously stirred at

TABLE I

Purijkation of KB cell DNA polymerases

il. Cytoplasmic polymerase

B. ?Juclear polymerases

1. Polymerase ?Jl

2. Polymerase N2

.- Fraction or step

Crude extract” Acid precipitation Ultracentrifugation DEAE Sephadex G-200 Phosphocellulose

Nuclear extract,b polymerase r\‘l

/ DEAE

\ uolrmerase ?i2

Phosphocell;lose Sephadex G-200 Isoelectricfocusing Phosphocellulose

Hydroxglapatite

V0hlIe Protein Activitr jpecific activity Yield

ml

433 33 44

5.5 35

5

300 339

50 2

3 2.5

74

13

*Bg

4,330

1,716 1,276

95 37

0.9

u?ds

14,263

18,262 16,495 4,279 3,486

1,470

units/mg %

3 (100) 11 12X

13 115 45 30 94 24.4

1,633 10.3

399 10,591 3i 146 3,115 21

146 10,669 73 (1OO)c 3.3 575 176 18.5 0.41 193 466 6.2 0.012 109 8,720 3.5

27 2,158 79 20.2

1.8 1,246 6!r” 11.6

a Crude extract was from 40 g of KB cells.

b Nuclear extract was from 2 x 1010 nuclei or approximately 100 g of KB cells. c In calculating yield of the nuclear polymerases, the DEAE-fractions were taken as 100%.

-

(1oo)c

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5029

4” for 4 hours. The viscous extract was dialyzed for 48 hours at 4” against three changes of 10 volumes each of 0.025 M po- tassium phosphate, pH 8.5. The bulk of the nuclear DNA was precipitated by this procedure and was removed by cen- trifugat.ion at 10,000 g for 5 min.

Polymerase Ni

DESE-cell&se Column Chromatography-Nuclear crude ex- t,ract was added to a DEAE-column (2.5 x 15 cm), and the column was washed with 0.025 M potassium phosphate, pH 8.5. The fraction of added polymerase activity that did not adsorb t,o the column was pooled and concentrated.

Phosphocellulose Column Chromatography-The concentrated enzyme fractions were adjusted to 0.15 M potassium phosphate, pH 7.2, and loaded on a phosphocellulose column (2.5 x 7.5 cm). The column was washed with 0.15 M potassium phosphate until all unadsorbed protein was removed. Enzyme activity was eluted either at 0.25 M with a 150-ml gradient of 0.15 M to 0.40 RI of potassium phosphate, pH 7.2, or in a single step at 0.26 M.

Fractions with enzyme activity were pooled and concentrated.

tract and has been routinely purified 500- to 1000-fold in 5 to 10% yield to a specific activity of 1600 to 2500. All of the data relating to polyrnerase C that are presented in this paper were obtained with the phosphocellulose fraction (Table I, Part A). Polymerase C is a more highly purified form of the major polym- erase activity we previously described in frozen KB cells (17), and like that activity, neither polymerase C enzyme activity nor protein from the phosphocellulose fraction enters agarose or polyacrylamide gels under nondenaturing conditions. In SDS acrylamide gels different phosphocellulose preparations of polymerase C exhibit two to four bands in the molecular weight range of 1.1 to 1.4 X 105. This size is in agreement with the molecular weight estimation (1.4 X 105) of polymerase C en- zyme activity by Sephadex G-200 gel filtration, but we have not yet established that any of the bands observed in SDS gels represent enzyme protein.

G-200 Gel Filtration-Gel filtration was carried out as de- scribed for polymerase C. Polymerase activity was eluted in fractions of about 38,000 molecular weight as indicated by pro- tein standards.

Isoelectricfocusing-Isoelectricfocusing was carried out as de- scribed for the LKB 8101 (110 ml) column in the LKB manual. Ampholytes in the pH 8 to 10 range were used at a concentration of 2%. Polymerase Nl was focused at pH 9.2. Fractions containing polymerase activity were pooled, adjusted to 20% ethylene glycol and stored above liquid nitrogen for at least 6 months without an appreciable loss of activity.

Nuclear Polymerase I-Two DNA polymerase activities have been identified in the KB nuclear extract. Polymerase Nl has been purified about 700-fold from the DEAE-fraction in 10 to 15% yield to a specific activity of 8720. The data reported herein for this enzyme were obtained with the fraction purified through the isoelectricfocusing step (Table I, Part 731). The extremely small quantities of purified polymerase Nl available to date have precluded a detailed study of its physical properties. However, on slab SDS polyacrylamide gel electrophoresis, the most purified fraction exhibits a single band of molecular weight 3.9 x 104. This size agrees with the estimated molecular weight (3.5 x 103 of polymerase Nl enzyme activity obtained by gel filtration.

A representative purification of polymerase Nl is shown in Table I, Part B.

Polymerase N2

DEAE-cellulose Column Chromatography-The nuclear crude extract was added to a DEBE-column (2.5 x 15 cm). The column was washed with 0.025 M potassium phosphate, pH 8.5, unt,il all unadsorbed polymerase activity was removed, and then developed with a linear gradient of 150 ml of 0.05 M to 0.2 M potassium phosphate, pH 8.5. A peak of polymerase activity was eluted at 0.15 M. Fractions containing this enzyme activity were pooled and concentrated.

Nuclear Polymeruse II-The second DNA polymerase activity in the nuclear extract, polymerase N2, is the least pure of the three KB enzymes The best preparations to date (Hydroxyl- apatite fraction, Table I, Part B2) have a specific activity of 692, although with exonuclease III-treated DNA template the specific activity increases to 1300 (see below). This represents a purification of about lo-fold from the DEAE-fraction in 5y0 yield. On SDS polyacrylamide gels, the best polymerase N2 fractions contain numerous bands. Only few of the properties of the hydroxylapatite fraction of this enzyme are included for comparative purposes in this report.

Properties of KB Polymerases

Phosphocellulose Column Chromatography-Concentrated en- zyme fractions were loaded on a phosphocellulose column (2.5 x 7.5 cm). The column was washed and developed as described above for polymerase Nl. Polymerase activity was eluted at 0.21 31 potassium phosphate. Fractions with enzyme activity were pooled and concentrated.

General Requirements-As shown in Table II, the three KB enzymes have identical pH optima and very similar divalent cation requirements. They differ in their K, values for deoxy- nucleoside triphosphate and strikingly in their apparent K, values for activated salmon sperm DNA. The effects of several

TABLE II Hydroxylapatite Column Chromatography-Concentrated en-

zyme IT-as diluted to 0.05 M potassium phosphate, pH 7.2, and applied to a hydroxylapatite column (1.5 x 2 cm). The column was washed until all unadsorbed protein was removed. Enzyme activity was eluted as a single peak at 0.25 M by a 60-ml linear gradient of 0.05 M to 0.40 M of the same buffer. Polymerase N2 was stored over liquid nitrogen for more than 6 months with- out loss of activity.

Some characteristics of KB DNA polymerases The assays were carried out with 500 ng of polymerase C, 25

ng of polymerase Nl and 1.4 pg of polymerase N2. The primer- template was activated salmon sperm DNA.

Polymerase c Polymerase Nl I I I

Polymerase N2

A representative purification of polymerase N2 is shown in Table I, Part B.

Purity of Polymerases

pH optimuma.. 9.2 9.2 9.2 Mg++ optimum. 10 mM 20 rnM 10 rnM

K,of dNTP.... 0.5 X 10-4M 1.G x 1O-4 M 0.23 x 10-4~ K, of DNA.. 50 pg per ml 6.7 pg per ml 15.2rgperml Isoelectric pH 5.6b 9.2

Cytoplasm&z Polymerase-A single DNA polymerase activity a The buffer was Tris-chloride. has been identified in the organelle-free cytoplasmic crude ex- b Enzyme activity collected in a precipitate at pH 5.6.

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5030

chemical inhibitors on the activities of polymerases C and Nl are presented in Table III. As is generally true for the DNA polymerases, pyrophosphate strongly inhibits both enzymes. Polymerase C is slightly inhibited by acridine orange and mark- edly inhibited by ethidium bromide, while neither of these inter- calating agents affects polymerase Nl activity at the concen- trations tested. Polymerase C appears to be very sensitive to p-hydroxymercuribenzoate, in contrast to the insensitivity of polymerase Nl. The enzymes also differ in their response to monovalent cations, polymerase C showing profound inhibi- tion in 0.15 M to 0.20 M salt, while polymerase Nl activity is partially stabilized by 0.20 M KCI. Thus, under standard assay conditions dNTP incorporation by polymerase Nl ceases after 30 min; in the presence of 0.20 M KU, incorporation proceeds for up to 3 hours (see Fig. 1).

1Vuclease Activities-The three KB polymerases demonstrate no detectable endonuclease activity when tested by a highly sensitive assay for nicking of supercoiled molecules of SV40 DNA (see “Experimental Procedure”). Polymerases C and N2 exhibit a low level of exonuclease activity (1 to 2yo of polym- erizing activity), while the most purified fraction of polymerase Nl appears to be free of exonuclease (<0.2% of polymerizing activity). However, ultrasensitive assays for 3’ + 5’ exonu- clease activity (28, 29) have not yet been carried out. It is im- portant to note that the gel filtration fraction of polymerase Nl (Table I, Part Bl), which is at a stage of purification compar- able to that reported by others (13) (i.e. 500 to 700 units per mg), contains an exonuclease activity that is equal to the polymerase activity. The nuclease activity is removed completely in the isoelectric focusing step, and we have not studied it further.

DNA-Primer-Template Requirements-None of the three KB enzymes incorporates dNTP in the absence of exogenous DNA primer-template. The ability of the KB enzymes and of E. coli

DNA polymerase I to use salmon sperm DNA after various types of modification by nucleases is shown in Table IV. In the absence of nuclease digestion, the three KB enzymes use the DNA primer-template poorly. In contrast, the E. coli en-

TABLE III E$ect of chemical inhibitors on polymerase C and polymerase Xi

Standard assay conditions were used with the additions indi- cated in the table. In experiments testing the effect of p-hydroxy- mercuribenzoate, p-mercaptoethanol was omitted from the incubation mix. The assays were performed with 2.6 ,.~g of pol- ymerase C and 50 ng of polymerase Nl. Activity, lOOy,, for pol- ymerase C was 5.2 nmoles per hour; for polymerase Nl: 0.42 nmoles per hour.

Addition

None. Sodium pyrophosphate

5rnM

Acridine orange 2O/*M ,.,. 50/x.

Ethidium bromide 10~M..........

20,uM ,... p-Hydroxymercuribenzoate

10pM ._.................._.,.

39pM .._............_..... NaCl0.15M KC1 0.20 M

-

Polymerase c

70 acfioi/y

100

31

X6 85

81 38

20 13 20 11

Polymerase Kl

yc ac/ioiiy

100

37

98 95

106 111

105 100 113

zyme uses the native template with 10 to 20 times the eficieilcy of the KB enzymes. Digestion with DNase I increases tem- plate efficiency by 40- to loo-fold for the KB enzymes, j-fold for the E. coli enzyme. A similar degree of digestion with micro- coccal nuclease abolishes template activity for all of the enzymes, demonstrating the absolute requiremeut of the polyineruses for 3’-hydroxyl termini. The introduction of gaps into native DNA with exonuclease III (4, 31) produces a primer-template that is used with different degrees of facility by the different enzymes. The ratios of activity with exonuclease III template to that with DNase I template for the four enzymes are: polym- erase Nl, 0.14; polymerase N2, 1.9; polymerase C, 0.45; and E. coli, 1.1. Thus, the preference of polymerase N2 for the exonuclease III template was unique among the KB polymerases.

The KB polymerases differ dramatically from the R. co/i en zyme in their ability to use poly[d(A-T)] as primer-template (Table IV). Polymerase C uses poly[d(A-T)] least well, and the limited extent of incorporation is similar whether both d-YIP and dTTP are provided as substrates or whether either of the dNTPs is present alone. Thus, polymerase C seems able to perform only single deoxynucleotide addition to the 3’.hydroxyl termini in this copolymer. Polymerases Nl and N2 show greater ability to use this primer-template, and incorporation is j-fold greater when both dATP and dTT1’ are present than when either is present alone; thus, these enzymes do appear to effect polym- erization with the copolymer. It is important to emphasize that none of the KU enzymes will incorporate dCTP or dGTP with poly[d(A-T)], thus demonstrating their fidelity in copying this primer-template.

The final entry in Table IV indicates that all three KB en zymes can utilize a polyribonucleotide primer. Incorporation of [3H]dAMP using poly(A).poly[d(T)5&, is 10 to 45yo of the incorporation observed with poly[d(A-T)]. We have not yet attempted to optimize conditions for the polg(A.dT) primer- template, and greater efficiency may be obtainable.

FIG. 1. Kinetics of incorporation of 4 dNTPs versus 1 dNTP by polymerase Nl. The reaction mixtures contained, in a total volume of 2 ml: 10 mM Tris-HCl, pH 9.2; 2 mM p-mercaptoethanol; 200 pg per ml of bovine serum albumin; 10 mM MgC12; 0.2 M KCl; 120 rg of activated salmon sperm DNA and 200 ng of polymerase Nl. One mixture (O---O) contained 200 PM each of dATP, dCTP, dCTP and dTTP and [zH]dTTP (0.04 Ci per mmole); the second (O---O) contained 200 PM dTTP only plus [3H]dTTP (0.04 Ci per mmole). Incubation was carried out at 37”, and 0.25ml aliquots were removed at the indicated times for deter- mination of [3H]dTMP incorporation. At 180 min (I ) an addi- tional 100 ng of polymerase Nl were added to each reaction vessel and incitbation continued for another 180 min.

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5031

Utilization of RNA-DNA Hybrid Templates--The ability of Nl as well as the E. coli enzyme, possess a similar activity which the three KB polymerases to use hybrid templates was tested is particularly well demonstrated with poly[d(A-T)]. Polym- under the conditions described by Ross et al. (32), Stavrianopou- erase C, which carries out only a limited single nucleotide addi- 10s et al. (33), and Goodman and Spiegelman (34). We used tion with this template, utilizes dADP as well as dATP, and both poly(A) .poly[d(T)3od and poly(A) .poly[d(T)12..~] at ap- incorporation is not increased by adding ATP to the incubation. propriate pH values, with KCI, and with either &In++ or iLlg# Polymerase Nl uses dADP more efficiently than the E. coli ea- as divalent cation. Under the conditions reported to be optimal zyme, and incorporation is enhanced either by adding ATP to for hybrid template use, polymerization with activated DNA the reaction or by preliminary incubation of polymerase Nl was poor; polymerase C, 15% and polymerase Nl, 22%, re- with substrates and ATP in the absence of template. With spectively, of the levels of incorporation observed under standard activated DNA primer-template, all three enzymes show low assay conditions. Reading of the polyribonucleotide strand but significant incorporation with dADP, and incorporation of the hybrid was at best 0.9% (polymerase C) and 0.47$ (polym- is stimulated 3- to 5-fold with each enzyme in the presence of erase Nl) of the reduced activity observed with activated DNA ATP. under the same conditions. Thus, the highly purified polym- Reaction of DNA Polymerases with a Single dNTP---The KB erases C and Nl cannot use a polyribonucleotide as template. polymerases require the presence of 4 dNTPs for maximal ac- Polymerase N2 has not yet been adequately tested. tivity with activated DNA primer-template. However, as with

Use of Deoxynucleoside Diphosphate as Substrate by DNA most eukaryote DNA polymerases, the KB enzymes demonst.rate Polymeruses-Miller and Wells (35) have reported the presence substantial incorporat.ion when only a single dNTP is present’. of nucleoside diphosphokinase-like activity in E. co& DNA We have studied this activity in more detail and have observed polymerase I. As shown in Table V, both polymerases C and that the extent of reaction with a single dNTP is a function both

TABLE IV

DNA template requirements of polymerases For the KB polymerases, assays were performed under standard conditions using 500 ng of polymerase C, 75 ng of polymerase Nl

or 1.5 pg of polymerase N2. E. coli polymerase I, 120 ng, was assayed under standard conditions for that enzyme (30). The primer- templates in the different experiments were 25 pg of salmon sperm DNA, treated as indicated in the table; or 25 pg of poly[d(A-T)];

or 16 nmoles (measured as dT nucleotide) of

F Template

Polymerase C

DNase I-treated native DNA.. Native DNA..............................

Denatured DNA.. Exo III-treated native DNA. _. Micrococcal nuclease-treated native DNA..

@mle/hr

697 8

22 313

0

poly[d(A-T)] . . . . . 29

pol~(A)~poly[d(T) I 5000 . . . . . . . . . . . . . . . . . . . . 13

dAMP incorporated

T Polymerase iv1 Polymerase N2 E. coli polymerase I

I- % (100)

(1.2) (3.2)

(45) 0

(4.2) (1.9)

@nole/hr % 605 (100)

5 (0.8) 22 (3.7) 87 (14.4)

0 0

776 18 89

1489 0

% @de/hr

(100) 662

(2.3) 125

(11.4) 28

(192) 717 0 0

121 (20) 390 (50.3) 22 (3.6) 46 (5.9)

1136

(lo:) (18.8)

(4.2) (108)

0

(172)

I - -

TABLE V

Use of deoxynucleoside diphosphate as substrate by DNA polymerases Standard incubation conditions were used for the KB polymer- of polymerase C, 25 ng of polymerase Nl and 600 ng of E. coli

ases and for E. coli polymerase I, respectively. Substrates, tem- polymerase I, for 1 hour at 37”. The labeled substrates (dADP plates (100 pg per ml) and addition of ATP (1 mM) were as indi- and dATP) were checked for purity by paper chromatography

cated in the table. When dADP was used as substrate in the (36). The chromatograms were cut into l-cm strips and counted absence of ATP, the other dNTPs were present at twice their in a liquid scintillation spectrometer. dATP contamination of

normal concentration. The assays were performed with 500 ng the tritiated dADP was <O.O7Yo.

Incorporation of [PH-IdAMP

Substrate Addition Template Preincubation of E. cdi DNA polymerase Nl’= polymerase I Polymerase c Polymerase Nl

$molelhr / @nole/hr

39 (100%) 121 (100%)

39 (loo%‘,) 29 (24%) 36 (92%) 50 (41%) ,

fmwle/hr

25 (21%) 84 (7’3Yoh)

$mmle/hr

5680 (100%)

615 WY,) 884 (16%)

246 (1OOYo) 195 W%‘,) 3309 (100%)

31 (12.5Yo) 4 (2.3%) 37 0.1%)

83 (33Yo) 22 WYO’,) 170 (5.2%)

MA-T)1

Activated DNA

dATP, dTTP None dADP, dTTP None dADP, dTTP ATP

dATP, dTTP, dGTP, dCTP None dADP, dTTP, dGTP, dCTP None dADP, dTTP, dGTP, dCTP ATP

a Preincubation was carried out at 37” for 60 min in the presence of all constituents of the assay except template. The tubes were heated at 100” for 2 min, and the polymerization was then begun by addition of template and fresh enzyme.

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5032

of the primer-template and of the duration of incubation. Thus, with activated DNA and E. coli polymerase I, the extent of incorporation with 1 dNTP versus 4 dNTPs was 4.5% after 5 min, 2% after 10 min, and <l% after 30 min, the usual incu- bation time. Under similar conditions the comparable values obtained with the KB enzymes after 10 min of reaction were polymerase C, 5 to 10%; polymerase N2, 1070; and polymerase Nl, 80%.

The kinetics of incorporation of 1 dNTP versus 4 dNTPs by polymerase Nl with activated DNA are depicted in Fig. 1. In contrast to E. coli polymerase I, in which incorporation of a sin- gle dNTP has reached completion by 5 min, the rate of incorpo- ration of a single dNTP by polymerase Nl is comparable to that observed in the presence of all 4 dNTPs during the first 10 min of incubation. Thereafter, the rate of reaction abruptly declines (3 to 5% of initial rate) and is unaffected by the addition of fresh enzyme at 180 min. With 4 dNTPs, incorporation is brisk during the initial 30 min of reaction and then proceeds at a re- duced but linear rate (20% of initial rate) for another 150 min. Addition of fresh enzyme at 180 min reproduces the initial reac- tion kinetics, and after 360 min of incubation, the extent of in- corporation with a single dNTP is only 15% of that attained in the complete reaction. The biphasic kinetics suggest that a rapid rate of initiation is followed by a slower rate of polymeriza- tion, the extent of which is limited by enzyme instability that has not yet been overcome.

By incubating increasing quantities of E. coli polymerase I with a limited amount of DNA primer-template for 5 min in the presence of a single dNTP, it is possible to achieve saturating levels of enzyme and to measure the number of 3’-hydroxyl ini- tiation sites by the extent of incorporation of the single deoxy- ribonucleotide (37). When the salmon sperm DNA prepara- tions used in this report were analyzed in this way, the extent of incorporation with native DNA was 11 pmoles/25 pg of DNA (2.6 x 1Ol3 3’-OH sites) ; with exonuclease III-treated DNA, 36 pmoles/25 pg of DNA (8.6 X lOi 3’-OH sites) ; and with DNase I-treated DNA, > 100 pmoles/25 pg (>2.4 X lOi 3’-OH sites). The total amount of single nucleotide incorporated by polymerase Nl (and the other KB enzymes) never exceeded, and was usually but a small fraction of the total available 3’-hy- droxyl termini.

Finally, we note that none of the KB polymerases demon- strates activity when assayed with d(pT)g under optimum con- ditions for terminal transferase (38), nor do they incorporate ribonucleotides in the presence or absence of dNTPs.

DISCUSSION

Several recent papers have noted the presence of distinct DNA polymerases in crude nuclear and cytoplasmic extracts of mam- malian cells (13-16), and Weissbach et al. (13) have reported some of the properties of three partially purified DNA polym- erases in extracts of frozen HeLa cells. The three enzymes we describe in this paper were obtained from freshly harvested human KB cells and resemble the HeLa cell polymerases in their distribution and major characteristics. The several reports are in agreement that a low molecular weight DNA polymerase can be isolated from nuclei and a high molecular weight DNA polymerase from the cytoplasm.

We have purified polymerases C and Nl to specific activities of 1600 to 2500 and 7500 to 8000, respectively, which compare favorably with those of the most highly purified preparations of E. coli DNA polymerase I (18) and II (5). The most purified fraction of polymerase Nl contains only a single protein band of

39,000 daltons by SDS-polyacrylamide gel electrophoresis, but because of the very limited amounts of enzyme available, the coincidence of enzyme activity and protein has not yet been es- tablished in nondenaturing gels. The best preparations of polymerase C, the polymerase we have described (17) in extracts of frozen KB cells, presents two to four bands in SDS gels in a molecular weight range (110,000 to 140,000) that is consistent with that of polymerase C enzyme activity as measured by gel filtration. As we previously reported, this enzyme does not enter polyacrylamide gels under a variety of nondenaturing conditions, for reasons not yet clear.

The third KB polymerase, polymerase N2, has only been purified to a degree comparable to that described by Weissbach et al. (13). This enzyme is similar in chromatographic behavior and molecular weight (as estimated by gel filtration) to polym- erase C, but our preliminary studies have revealed a number of enzymatic properties that distinguish it from the cytoplasmic polymerase. Thus, in comparison to polymerase C, polymerase N2 shows a higher affinity for DNA, preference for exonuclease III-treated DNA primer-template and relatively greater ability to use poly[d(A-T)]. Nonetheless, until the polymerase N2 activity is further purified, its identity as a unique, third KB polymerase cannot be proved. It is of interest, however, that all three KB enzymes differ significantly from a highly purified DNA polymerase from rat liver (39), suggesting that there may be distinctive differences between the DNA polymerases of human and nonhuman mammalian cells.

None of the three KB polymerases contains detectable endo- nuclease activity. Polymerases C and N2 have low levels of an exonuclease activity that has not been characterized, but we have not been able to detect exonuclease activity in the most purified fraction of polymerase Nl using conventional assays. Although cytoplasmic extracts of KB cells contain a hybrid template activ- ity (unpublished experiments) similar to that recently described (32, 34) in mammalian cells, the activity is not associated with any of the three KB DNA polymerases. This finding is in agree- ment with the recent HeLa cell study reported by Fridlender et al. (40). The KB enzymes differ from E. coli polymerase I in their inability to copy a polyribonucleotide strand under the assay conditions of Karkas et al. (41). Like the E. coli enzyme, both polymerases C and Nl can use dADP in place in added dATP as substrate for polymerization, and utilization is en- hanced in the presence of ATP. Thus, this nucleoside diphos- phokinase-like activity may prove to be a general property (or contaminant) of both prokaryotic and eukaryotic DNA polym- erases.

A particularly vexing problem with most reported mammalian DNA polymerases (see References 17 and 42) has been their apparently disproportionate reactivity in the presence of only a single dNTP. By measuring the number of available 3’-hy- droxyl termini in our DNA primer-templates, we have now clearly demonstrated that the amount of single nucleotide incorporated is always substantially less than the number of initiation sites available for single base addition. This should eliminate allegations that mammalian DNA polymerase prepara- tions are necessarily contaminated by a terminal transferase activity that seems, in fact, to be limited uniquely to the thymus gland (43).

Like all other DNA polymerases, the three KB enzymes have an obligatory requirement for 3’-hydroxyl initiation sites. In their utilization of templates, the KB enzymes resemble E. coli polymerase II (31) in showing an apparent requirement for gaps and an inability to effect polymerization at nicks. This conclu-

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5033

sion is based on the following observations: Native salmon sperm DNA is almost inactive as template, and yet, as determined with E. coli polymerase I, it contains a large number of 3’-hy- drosyl termini (2.6 X 10i3/25 pg of DNA). The template can be activated 20- to 40.fold by digestion with exonuclease III, a treatment that converts nicks to gaps, but leads to only a modest increase in 3’.hydroxyl termini (8.6 X 1013/25 pg of DNA). The DNA is also markedly activated by extensive digestion with DBase I (20% of the DNA rendered acid-soluble) which increases the number of 3’.hydroxyl termini >lO-fold, but more signifi- cantly generates a large number of small gaps (31). The re- quirement for gaps most likely reflects the absence in the KB enzymes of the 5’.exonuclease activity of E. coli polymerase I (44), and inability to displace the 5’-strand ahead of the grow- ing point in the absence of nucleolytic digestion (29). The rela- tive preference of the three KB polymerases for DNase I-treated versus exonuclease III-treated templates varies (Table IV). The most striking discrimination is exhibited by polymerase Nl, whose preference for DNase I-treated template may reflect an inability to traverse long gaps. This raises the question of whether such gaps might be more effectively utilized in the pres- ence of a DNA-stabibzing protein such as the phage T4 gene 32 product (45, 46) and suggests a possibly convenient assay sys- tem with which to seek such a specific protein in KB cells.

Finally, we note that all of the KB polymerases can utilize poly(h) as primer for the incorporation of dAMP on a poly(dT) template. Although we have not yet studied this type of reac- tion in detail, the property is noteworthy in view of recent reports linking DNA polymerization to RNA polymerization in vitro (41) and in vivo (47-49).

Acknowledgments-We acknowledge the expert assistance of NIrs. M. Sedwick in preparing cell cultures, and we thank I. R. Lehman, A. Kornberg, and D. Brutlag for critical review of the manuscript.

REFERENCES

1. DELUCIA, P., AND CAIRNS, J. (1969) iVatul‘e 224, 1164 2. Moses, 1~. E., AND RICHARDSON, C. C. (1970) Biochem. Bio-

phys. Res. Commun. 41, 1557 and 1565 3. KXI~PERS, R. (1970) Nature 228, 1050 4. KORNBERG, T., AND GEII’TER, M. L. (1971) Proc. iVat. Acad.

Sci. U. S. A. 68, 761 5. WICKNER, R. B., GINSBERG, B., BERKO~~ER, I., AND HURXVITZ,

J. (1972) J. Biol. Chem. 247, 489-497 6. GEFTER, M. L., HIROTA, Y., KORNBERG, T., WECHSLER, J. A.,

AND Ba~ivoux, C. (1971) Proc. Nat. Acad. Sci. U. S. A. 68, 3150

7. RAMA REDDY, G. V., GOULIAN, M., AND HENDLER, S. S. (1971) Nature New Biol. 234, 286

8. BOLLUM, F. J. (1960) J. Biol. Chem. 236, 2399 9. YONEDA, M.. AND BOLLUM. F. J. (1965) J. Biol. Chem. 240.

3385-3391 ’ I

10. KORNBERG, A. (1969) Science 163, 1410 11. KALF, G. F., AND CH’IH, J. J. (1968) J. Biol. Chem. 243, 4904-

4916 12. MEYER, R. R., AND SIMPSON, M. V. (1970) J. Biol. Chem. 246,

3426-3435 13. WIXSSBACH, A., SCHLABACH, A., FRIDLENDER, B., AND BOLDEN,

A. (1971) Xature New Biol. 231,167

14. BARIL, E. F., BROX-X, 0. E., JCNI~INS, M. D., -44~‘~ L.UZLO, J. (1971) Biochemistry 10, 1971

15. HOWB, R., AND WANG, T. Y. (1969) Arch. Biochem. Biophys. 133, 238

16. CHANG, L. M. S., AND BOLLUM, F. J. (1971) J. Biol. Chrm. 246, 5835-5837

17. GREENE, R., ASD KORN, D. (1970) J. Biol. Chem. 246,2,X-261 18. JOVIX, T. M., ENGLUND, P. T., AND BERTSCH, L. L. (1969) J.

Biol. Chem. 244, 2996-3008 19. APOSHIAA-, H. V., I\ND KORNBEILG, A. (1962) J. Biol. Chem.

237,519-525 20. WIXISSBACH, A., AND KORN, D. (1963) J. Biol. Chem. 238, 3383-

3389 21. FRIEDBERG, E. C., .\ND CLAYTON, D. A. (1972) i\‘ature 237, 99-

100 22. VINOQRAD, J., LEHOX-ITZ, J., RADLOFF, R., WATSON, R., .IND

LAIPIS, I?. (1965) Proc. Nat. Acad. Sci. U. S. A. 63, 1104 23. CLAYTON, D. ,1., AND VINOGRAD, J. (1967) iVature 216,652-657 24. LOWTRY, 0. H., ROSEHROUGH, N. J., FARR, A. L., .ZWD RAN-

DALL, R. J. (1951) J. Biol. Chem. 193, 265-275 25. GROVES, W. E., DAYIS, F. C., JR., AND SELLS, B. H. (1968)

Anal. Biochem. 22, 195 26. REID, M. S., AND BIELESI~I, R. L. (1968). Anal. Biochcv~. 22,

374 27. LAEMMLI, U. K. (1970) Nature 227,680-685 28. BRUTL.~G, D., BND KORNB~~RG, A. (1972) J. BioT. Cherry. 247,

241-248 29. SETLOW, P., BRUTLAG, D., AND KORNBERG, A. (1972) J.

Biol. Chem. 247, 224-231 30. RICHARDSON, C. C., SCHILDBRIUT, C. L., APOSHIAN, H. V.,

AND KORNBXILG, A. (1964) J. Biol. Chem. 239, 222 31. WICKNER, R. B., GINSI~I~RG, B., AND HU~XITZ, J. (1972) J.

Biol. Chem. 247, 498-504 32. Ross, J., SCOLNICK, E. M., T~DARO, G. J., AND AARONSON, S.

A. (1971) Safuye Avew Biol. 231. 163 33. STAVRIAN~POULOS, J. G., KARK~S, J. D., AND CHARG.\FF, E.

(1971) Proc. -\-at. Acad. Sci. U. s. A. 68, 2207 34. GOODMAN. N. C.. .\ND SPIEGELMAN. S. (1971) Proc. Nat. Acad.

I ~

Sci. U. S. A. 66, 2203 35. MILLE:R, L. K., AND WELLS, R. D. (1971) Proc. Nat. Acad. Sci.

U. S. A. 66, 2298 36. Schwarz/Mann Radiochemical Catalog, 197O/Yl, page 6, solvent

system B 27. 37. ADLER, J., LEHX~N, I. R., BESSMAN, M. J., SIMMS, E. S., .~ND

KORNBERG, A. (1958) Proc. Nat. Acad. Sci. U. S. A. 44, 641 38. CHANG, L. M. S., AND BOLLUM, F. J. (1971) J. Biol. Chem. 246,

909-916 39. BJCRGF,R, H., HUANG, R. C. C., AND IRVIN, J. L. (1971) .I. Biol.

Chem. 246, 7275 40. FRIDLENDER, B., FRY, M., BOLDEN, A., .ZND WEISSBXH, A.

(1972) Proc. Nat. Acad. Sci. U. S. A. 69,452 41. K~RKA~, J. D., STAVRIANOPOULOS, J. G., END CHARGAFF, E.

(1972) Proc. -Vat. Acad. Sci. U. S. A. 69.398 42. GO‘ULIAN, M. (1971) Ann. Rev. Biochem. 40, 855 43. CHANG, L. M. S. (1971) Biochem. Biophys. Res. Commun. 44,

124 44. DEUTECHER, iU. P., BND KORNBERG, A. (1969) J. Biol. Chem.

244, 3029 45. ALBERTS, B. XI., .~SD FREY, L. (1970) Nature 227, 1313 46. HUBERMAN, J. -4., KORNBERG, A., AND ALBCRTS, B. M. (1971)

J. Mol. Biol. 62,39 47. DOVE, W. F., HARGROVE, E., OHASHI, M., HAUGLI, F., AND

GUHA, A. (1969) Jap. J. Genet. 44, Suppl. 1,ll 48. BRUTLAG, D., SCHEKMAN, R., AND KORNBERG, A. (1971) Proc.

Nat. Acad. Sci. U. S. A. 68, 2826 49. BAZZIC~LUPO, P., AND TOCCHINI-VALENTINI, G. P. (1972)

Proc. Nat. Bead. Sci. U. S. A. 69, 298

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. David Sedwick, Teresa Shu-Fong Wang and David KornPolymerases from Human KB Cells

Purification and Properties of Nuclear and Cytoplasmic Deoxyribonucleic Acid

1972, 247:5026-5033.J. Biol. Chem. 

  http://www.jbc.org/content/247/16/5026Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/247/16/5026.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 26, 2018

http://ww

w.jbc.org/

Dow

nloaded from