THE Joumn~ OF BIOLOGICAL CHEMISTRY Vol 253, No 10, Issue of May 25, pp. 3400-3401, 1978 Prrnied in U S.A. An Endodeoxyribonuclease of Human KB Cells PURIFICATION AND PROPERTIES OF THE ENZYME* (Received for publication, June 23, 1977, and in revised form, January 16, 1978) TAKASHI TSURUO,$ MAX ARENS, R. PADMANABHAN,§ AND MAURICE GREENS From the Institute for Molecular Virology, St. Louis University School of Medicine, St. Louis, Missouri 63110 An endodeoxyribonuclease has been purified 750-fold from human KB cells. The purified endonuclease requires Mgs+ for maximum activity; Mn’+ was less than half as active and Ca’+ inhibited the reaction. The optimum pH is 8.8 in Tris- HCI and the optimum buffer concentration is 10 mM. KC1 (and NaCl), -SH-reacting reagents, and tRNA strongly inhibit the reaction. An apparent molecular weight of 54,000 was determined by sedimentation in a glycerol gradient. The purified endonuclease cleaved native, double-stranded adenovirus 2 DNA, and the reaction proceeded stepwise during the initial stage of degradation by cleavage of the DNA substrate in half, then in half again, etc. At longer digestion times, single strand scissions were detected. RNA was not a substrate for the enzyme. Poly(dG) . poly(dC) was susceptible but poly(dA) poly(dT) was resistant to degrada- tion. Hydrolysis of adenovirus 2 DNA yielded double- stranded polynucleotides containing 5’-phosphoryl and 3’- hydroxyl termini with short, single-stranded regions pre- sumably at the ends. More than 50% of the product of a limit digest had a chain length greater than 35 to 40 nucleotides. Analysis of the 5’ and 3’ end groups of the digestion products indicated a preference for the site of the enzymatic cleavage; thymidylic acid residues were present at the 5’ end and deoxyguanosine residues at the 3’ end, each with a fre- quency of 40 to 50%. When cultured human KB cells are infected with human adenovirus type 2 (Adz), I the virions are partially uncoated in the cytoplasm and transported to the nucleus where uncoating * This work was supported by United States Public Health Service Grants AI-01725 and CA 17006 and Contract NO1 CP 43359 within the Virus Cancer Program of the National Cancer Institute and in part by Public Health Service Grants CA 20369 (R. P.) and AI 13540 (M. A.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “‘uduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Present address, Cancer Chemotherapy Center, Tokyo, Japan. 0 Recipient of Research Career Development Award 7K04-CA- 00235 of the National Cancer Institute. Present address, Department of Biological Chemistry, University of Maryland Medical School, Baltimore, Md. 21201. II Recipient of Research Career Development Award 5K6-AI-4739 of the National Institutes of Health. To whom reprint requests should be addressed. ’ The abbreviation used is: Ad2, human adenovirus type 2. is completed (1, 2). Three possible fates then exist for parental Ad2 DNA: (a) to serve as template for host cell RNA polym- erase II (11, (b) integration into host cell DNA, (c) degradation by cellular or possibly viral enzymes. Process a occurs in human KB cells which are productively infected by Ad2. With regard to b and c, Doerfler and co-workers (3) have reported that Ad2 DNA in infected KB cells is cleaved endonucleolyti- tally in the cytoplasm and nucleus and that about 20% of nuclear viral DNA becomes associated with cellular DNA. In order to elucidate the participation of endonucleases in these processes, we have initiated a study of the endonucleases of AdB-infected and -uninfected KB cells. A few systems are known in which mammalian endonucleases have been char- acterized. Multiple deoxyribonuclease activities were found in HeLa cells (4). Endonucleases that hydrolyze RNA and dena- tured DNA but not native DNA have been partially purified from human lymphoblasts and from hamster cells transformed by polyoma virus (5, 6). In this paper, we report the purifica- tion and properties of an endonuclease from human KB cells. EXPERIMENTAL PROCEDURES Growth ofKB Cells and Infection with Ad2 A clonal KB cell line was grown in suspension in Eagle’s minimal essential medium containing 5% horse serum (7). The cells were collected by centrifugation and washed with phosphate-buffered saline. The packed cells were stored at -80°C until used. Ad2 DNA Unlabeled or 3H-labeled Ad2 DNA was extracted from CsCl- purified virions as described previously (8). Labeling was accom- plished by growth of infected cells in media containing 1.0 @X/ml of f3Hlthymidine (New England Nuclear, 0.5 Ci/mmol). Neutral and Alkaline Sucrose Density Gradient Centrifugation of Ad2 DNA Samples for analysis were mixed with equal volumes of 0.6 M NaOH, 1 M NaCl, and 20 rnM EDTA or 20 mM Tris-HCl (pH 8.1), 0.4 M NaCl, 2 rnM EDTA, and 0.2% sodium dodecyl sulfate for alkaline or neutral sucrose gradient centrifugation, respectively. After incu- bation at 37°C for 5 min, the samples were layered onto a 5 to 20% (w/v) alkaline sucrose gradient containing 0.3 M NaOH, 0.5 M NaCl, and 10 rnM EDTA or a 5 to 20% (w/v) neutral sucrose gradient containing 10 rnM Tris-HCl (pH 8.1), 0.2 M NaCl, 1 mM EDTA, and 0.1% sodium dodecyl sulfate. The gradients were centrifuged for 2l12 h at 40,000 rpm in the Beckman SW 50.1 rotor or for 15 h at 23,000 rpm in the Beckman SW 41 rotor at 20°C. Fractions were collected from the bottom and radioactivity was determined by counting in Aquasol (New England Nuclear). 3400 by guest on August 6, 2019 http://www.jbc.org/ Downloaded from
Transcript
THE Joumn~ OF BIOLOGICAL CHEMISTRY
Vol 253, No 10, Issue of May 25, pp. 3400-3401, 1978
Prrnied in U S.A.
An Endodeoxyribonuclease of Human KB Cells PURIFICATION AND PROPERTIES OF THE ENZYME*
(Received for publication, June 23, 1977, and in revised form, January 16, 1978)
TAKASHI TSURUO,$ MAX ARENS, R. PADMANABHAN,§ AND MAURICE GREENS
From the Institute for Molecular Virology, St. Louis University School of Medicine, St. Louis,
Missouri 63110
An endodeoxyribonuclease has been purified 750-fold from human KB cells. The purified endonuclease requires Mgs+ for maximum activity; Mn’+ was less than half as active and Ca’+ inhibited the reaction. The optimum pH is 8.8 in Tris- HCI and the optimum buffer concentration is 10 mM. KC1 (and NaCl), -SH-reacting reagents, and tRNA strongly inhibit the reaction. An apparent molecular weight of 54,000 was determined by sedimentation in a glycerol gradient. The purified endonuclease cleaved native, double-stranded adenovirus 2 DNA, and the reaction proceeded stepwise during the initial stage of degradation by cleavage of the DNA substrate in half, then in half again, etc. At longer digestion times, single strand scissions were detected. RNA was not a substrate for the enzyme. Poly(dG) . poly(dC) was susceptible but poly(dA) poly(dT) was resistant to degrada- tion. Hydrolysis of adenovirus 2 DNA yielded double- stranded polynucleotides containing 5’-phosphoryl and 3’- hydroxyl termini with short, single-stranded regions pre- sumably at the ends. More than 50% of the product of a limit digest had a chain length greater than 35 to 40 nucleotides. Analysis of the 5’ and 3’ end groups of the digestion products indicated a preference for the site of the enzymatic cleavage; thymidylic acid residues were present at the 5’ end and deoxyguanosine residues at the 3’ end, each with a fre- quency of 40 to 50%.
When cultured human KB cells are infected with human adenovirus type 2 (Adz), I the virions are partially uncoated in the cytoplasm and transported to the nucleus where uncoating
* This work was supported by United States Public Health Service Grants AI-01725 and CA 17006 and Contract NO1 CP 43359 within the Virus Cancer Program of the National Cancer Institute and in part by Public Health Service Grants CA 20369 (R. P.) and AI 13540 (M. A.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “‘uduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$ Present address, Cancer Chemotherapy Center, Tokyo, Japan. 0 Recipient of Research Career Development Award 7K04-CA-
00235 of the National Cancer Institute. Present address, Department of Biological Chemistry, University of Maryland Medical School, Baltimore, Md. 21201.
II Recipient of Research Career Development Award 5K6-AI-4739 of the National Institutes of Health. To whom reprint requests should be addressed.
’ The abbreviation used is: Ad2, human adenovirus type 2.
is completed (1, 2). Three possible fates then exist for parental Ad2 DNA: (a) to serve as template for host cell RNA polym- erase II (11, (b) integration into host cell DNA, (c) degradation by cellular or possibly viral enzymes. Process a occurs in human KB cells which are productively infected by Ad2. With regard to b and c, Doerfler and co-workers (3) have reported that Ad2 DNA in infected KB cells is cleaved endonucleolyti- tally in the cytoplasm and nucleus and that about 20% of nuclear viral DNA becomes associated with cellular DNA.
In order to elucidate the participation of endonucleases in these processes, we have initiated a study of the endonucleases of AdB-infected and -uninfected KB cells. A few systems are known in which mammalian endonucleases have been char- acterized. Multiple deoxyribonuclease activities were found in HeLa cells (4). Endonucleases that hydrolyze RNA and dena- tured DNA but not native DNA have been partially purified from human lymphoblasts and from hamster cells transformed by polyoma virus (5, 6). In this paper, we report the purifica- tion and properties of an endonuclease from human KB cells.
EXPERIMENTAL PROCEDURES
Growth ofKB Cells and Infection with Ad2
A clonal KB cell line was grown in suspension in Eagle’s minimal essential medium containing 5% horse serum (7). The cells were collected by centrifugation and washed with phosphate-buffered saline. The packed cells were stored at -80°C until used.
Ad2 DNA
Unlabeled or 3H-labeled Ad2 DNA was extracted from CsCl- purified virions as described previously (8). Labeling was accom- plished by growth of infected cells in media containing 1.0 @X/ml of f3Hlthymidine (New England Nuclear, 0.5 Ci/mmol).
Neutral and Alkaline Sucrose Density Gradient Centrifugation of Ad2 DNA
Samples for analysis were mixed with equal volumes of 0.6 M
NaOH, 1 M NaCl, and 20 rnM EDTA or 20 mM Tris-HCl (pH 8.1), 0.4 M NaCl, 2 rnM EDTA, and 0.2% sodium dodecyl sulfate for alkaline or neutral sucrose gradient centrifugation, respectively. After incu- bation at 37°C for 5 min, the samples were layered onto a 5 to 20% (w/v) alkaline sucrose gradient containing 0.3 M NaOH, 0.5 M NaCl, and 10 rnM EDTA or a 5 to 20% (w/v) neutral sucrose gradient containing 10 rnM Tris-HCl (pH 8.1), 0.2 M NaCl, 1 mM EDTA, and 0.1% sodium dodecyl sulfate. The gradients were centrifuged for 2l12 h at 40,000 rpm in the Beckman SW 50.1 rotor or for 15 h at 23,000 rpm in the Beckman SW 41 rotor at 20°C. Fractions were collected from the bottom and radioactivity was determined by counting in Aquasol (New England Nuclear).
Assay 1 -This assay measures the extent of degradation of native Ad2 DNA (31 S) by zonal centrifugation in a neutral sucrose density gradient. The standard reaction mixture (50 ~1) contained 20 rn~ Tris-HCl (pH 8.81, 4 rn~ MgCI,, 10 fig of bovine serum albumin, 3 /*g of Ad2 l:‘HlDNA (45,000 cpm), and enzyme. The reaction was incubated for 30 min at 37”C, and terminated by the addition of 50 ~1 of an ice cold solution containing calf thymus DNA (60 pg) and bovine serum albumin (150 pg). The mixture was centrifuged in a neutral sucrose gradient as described above. In routine assays, the gradient was fractionated from the bottom into 10 fractions and the radioactivity of each fraction was counted in Aquasol. Native intact Ad2 DNA appeared in Fractions 2 through 4 (more than 80% of it appeared in Fraction 3). The number of breaks introduced into Ad2 DNA was determined from the centrifugation pattern in a manner similar to that described by Rhoades and Meselson (9). The average molecular weight of degraded DNA in each fraction was calculated theoretically according to Burgi and Hershey (10) and the number of breaks per Ad2 DNA genome was estimated. The total number of double strand breaks introduced in 3 /*g of native Ad2 DNA is 7.8 x lOl”.JFiNi, where 7.8 x 10”’ is the number of genome molecules in 3 pg of Ad2 DNA, Fi is the fraction of total radioactivity found in Fraction i, and Ni is the calculated number of breaks per genome of DNA found in Fraction i. A unit of endonuclease is defined as the amount causing 7.8 x 10” breaks in 3 fig of Ad2 DNA (i.e. with this definition, an average of 10 breaks are introduced in each molecule of Ad2 DNA). The activity of endonuclease was then calculated from each centrifugation pattern as follows.
7.8 x IO”‘. [FiNi I_ J
7.8 x 10” = 0.1 FiNi
.i After the enzyme reaction, the products could be kept for at least 1 week at -20°C after addition of calf thymus DNA and bovine serum albumin without significant change in the centrifugation pattern. The dilution of enzyme was carried out in a solution consisting of 50 mM Tris-HCl (pH 7.81, 0.1 M KCl, 1 rnM dithiothreitol, and 20% (v/v) ethylene glycol.
Assay 2 -This method measures the increased template activity of native calf thymus DNA for Escherichia coli DNA polymerase I after treatment with the endonuclease fraction. The reaction mix- ture (50 ~1) contained 30 rnM Tris-HCl (pH 8.51, 4 mM MgCl,, 10 pg of bovine serum albumin, 20 fig of native calf thymus DNA (Sigma, type I, 1 mg/ml, stored in 50% glycerol at -2O”C), and enzyme. After incubation at 37°C for 60 min, a mixture (75 ~11 containing 66.6 mM potassium phosphate (pH 7.3), 13.3 mM MgCl,, 1.33 mM 2-mercapto- ethanol, 133 FM each of dATP, dGTP, dCTP, 13H1dTTP (0.1 Cii mmol), and 0.1 unit of E. coli DNA polymerase I (P-L Biochemicals, Inc.) was added. After incubation for an additional 30 min at 37”C, aliquots of 100 ~1 were placed on DE81 filter paper discs (Whatman, 25 mm diameter) and washed six times with 5% Na,HPO,, twice with water, and once with ethanol. Radioactivity on the dried disc was counted in a toluene-based scintillation mixture. Routine en- zyme assays of column effluents or gradient fractions were carried out by this method. Using this assay, a unit of endonuclease is defined as the amount causing the additional incorporation of 1 nmol of 13HldTTP into DNA. Linear incorporation of 13HldTTP was observed with up to 0.8 unit of enzyme. Approximately 1.5 x lo4 cpm were incorporated into DNA when calf thymus DNA was not treated with endonuclease while an additional lo4 cpm of lSH]d’I”I’P was incorporated into DNA with 0.8 unit of enzyme.
5’ and 3’ End Group Analysis ofEndonuclease Digestion Products
The specificity of cleavage by the endonuclease was determined by analysis of the nucleotide residues at the 5’ and 3’ ends of the endonuclease digests. For 5’ end group analysis, the termini in the Ad2 DNA limit digest were dephosphorylated with alkaline phos- phatase, the 5’-OH end labeled with ly-32P1ATP and polynucleotide kinase, and ““P-labeled 5’-mononucleotides identified after release by treatment with venom phosphodiesterase. The incubation mix- tures (2 ml) containing 50 pg of Ad2 DNA, 20 mM Tris-HCl (pH 8.5), 4 rnM MgCl,, 0.2 mg/ml of bovine serum albumin, and 38 units of endonuclease were incubated for 30 min at 30°C. After the addition of 8 pmol of EDTA, 100 pmol of NaCl, and 0.9 unit ofE. coli alkaline phosphatase (Worthington, electrophoretically pure), the reaction mixture was incubated further at 60°C for 1 h. The reaction was terminated by the addition of neutralized phenol. After thorough mixing and centrifugation, the upper aqueous layer was dialyzed
against 0.01 M Tris-HCl (pH 7.4) containing 1 rnM EDTA. For the polynucleotide kinase reaction (111, 2.5 pg of the dephosphorylated Ad2 DNA in a solution containing 67 mM NaCl, 67 mM Tris-HCl (pH 7.61, 10 mM MgCl,, 10 mM dithiothreitol, and 22 prv~ [Y-~“PJATP were incubated with 1.5 units of the enzyme at 37°C for 3 h. The reaction was stopped by the addition of trichloroacetic acid to 5% and 50 pg of salmon sperm DNA. After centrifugation at 800 x g for 10 min, the DNA precipitate was dissolved in 0.2 N NaOH and reprecipitated twice with trichloroacetic acid. The precipitated DNA was extracted twice with ether and dissolved in 0.25 ml of 0.1 N NH,OH. The solution was neutralized with CO, gas, incubated with pancreatic deoxyribonuclease (5 pg) and 5 rnM MgCl, for 2 h, dried in a desiccator, and taken up in 40 ~1 of 30 mM Tris-acetate (pH 7.8) and 3 mM magnesium acetate. After digestion with 0.7 pg of snake venom phosphodiesterase (purified by pH 3.5 treatment (1211, the resulting 5’-mononucleotides were separated and analyzed as de- scribed (13).
The identification of the 3’ end group was carried out by a modification of the method of Roychoudhury et al. (141 using calf thymus deoxynucleotidyl terminal transferase and 1 a-:?‘PlCTP to label the 3’-terminal nucleotide with %P. The conditions for digest- ing Ad2 DNA with endonuclease were the same as described for 5’ end group analysis. To 900 pmol of 1 a-““PICTP dried in a siliconized glass tube were added 100 ~1 of Ad2 DNA endonuclease digest (12.7 pgl and 11 pg of 10 strength cacodylate buffer (1.4 M potassium cacodylate, 0.3 M Tris base, pH 7.61, 10 mM CaCl,, 1 mM dithiothrei- tol, and 1 ~1 of terminal transferase (1 mg/ml of solution; specific activity, 5000 units/mg). The mixture was incubated at 37°C for 3 h (the pH drops from 7.6 to 7.01, and the reaction was terminated by the addition of 0.4 ml of 0.5 N KOH. The tube was maintained at room temperature for 18 h to remove the second added ribonucleo- side, neutralized with 0.2 ml of 1 N HCl, and the pH adjusted to about 8.0 with 50 ~1 of 0.5 M Tris-HCl (pH 8.5). The terminal transferase addition products were then dephosphorylated by incu- bation with 0.43 unit ofE. coli alkaline phosphatase at 45°C for 2 h, the reaction terminated with 0.75 ml of neutralized phenol, and the aqueous layer extracted three times with ether. Ether was removed by evaporation and the 3’-labeled endonuclease digestion products were precipitated by trichloroacetic acid after the addition of carrier calf thymus DNA. The precipitated material was purified as de- scribed above for 5’ end group analysis, dissolved in 0.2 ml of 0.05 N NH,HCO,, (pH 8.0) containing 5 nmol of CaCl,, and 3.5 fig of micrococcal nuclease (Sigma Chemical Co., specific activity of 100 units/mg) were added. After incubation at 37°C for 1 h, the sample was taken to dryness in a desiccator, and taken up in a solution (30 ~1) containing 28 mM potassium phosphate (pH 6.21, 2.4 mM EDTA, 0.03% Tween 80, 25 nmol each of dAp, dTp, dGp, and dCp, and about 3 Kg (0.06 unit) of spleen phosphodiesterase (the enzyme from Worthington Biochemicals was purified further by passage through a phosphocellulose column). Enzyme digestion was carried out at 37°C for 12 h. The separation of the resulting “LP-labeled 3’-mono- nucleotides was performed as described by Wu (13).
Purification of Endonuclease
The following operations were carried out at 0-4°C. Assay 2 was used to measure endonuclease activity. The protein content of each fraction was determined by the method of Lowry et al. (15) with bovine serum albumin as standard.
Preparation of Extract- Packed cells (35 g; approximately 2 g of packed cells were obtained per liter of suspension culture) were suspended in 170 ml of 50 mM potassium phosphate (pH 7.4) containing 10 rnM 2-mercaptoethanol. The cells were disrupted by sonication in a sonic oscillator (Fisher Scientific Co., model 300) at full power until more than 98% of the cells were broken (usually 2 mini as monitored by phase contrast microscopy, and debris was removed by centrifugation at 10,000 x g for 15 min. The crude extract was then centrifuged for 2 h at 20,000 rpm in the Beckman SW 27 rotor and the clarified supernatant was recovered (Fraction I: extract).
Ammonium Sulfate Precipitation - EDTA and 2-mercaptoethanol were added to Fraction I to final concentrations of 2 and 20 mM, respectively. Ammonium sulfate (SchwarzlMann, ultrapure) was added with continuous stirring to 35% saturation. After 30 min, the solution was centrifuged at 10,000 x g for 20 min and the superna- tant was recovered. The supernatant was brought to 55% saturation with ammonium sulfate and placed in the cold for 30 min. The resulting precipitate was collected by centrifugation at 10,000 x g for 30 min, dissolved in 30 ml of 50 mM potassium phosphate (pH
7.01, 10 rn~ 2.mercaptoethanol, and 10% (v/v) ethylene glycol (DEAE buffer) and dialyzed against 2 liters of DEAE buffer for 3 h. A small precipitate was removed by centrifugation and supernatant was recovered (Fraction II: ammonium sulfate).
DEAE-cellulose Chromatography - Fraction II was applied to a column of DEAE-cellulose (Whatman DE23, 2.0 x 23 cm) which had been previously equilibrated with DEAE buffer. After washing the column with 100 ml of DEAE buffer, elution was performed with 300 ml of a linear gradient ranging from 0 to 0.7 M KC1 in DEAE buffer. The flow rate was 20 ml/h and 5.ml fractions were collected. Half of the DNA synthesis stimulating activity, as measured by Assay 2 (hereafter referred to as endonuclease activity), was eluted without adsorbing to DEAE-cellulose. The activity that eluted between 0.20 and 0.38 M KC1 was pooled and dialyzed against 2 liters of 50 rn~ potassium phosphate (pH 6.81, 10 rn~ 2-mercaptoethanol, and 15% (v/v) ethylene glycol for 4 h (Fraction III: DEAE-cellulose).
Phosphocellulose Chromatography - Fraction III was adsorbed onto a column (1.0 x 23 cm) of phosphocellulose (Whatman P-11) previously equilibrated with 50 nm potassium phosphate (pH 6.81, 10 rn~ 2-mercaptoethanol, and 10% (v/v) ethylene glycol (P buffer). The column was washed with 60 ml of P buffer and eluted with 120 ml of a linear gradient ranging from 0 to 0.7 M KC1 in P buffer. The flow rate was 15 ml/h and 3-ml fractions were collected. Two peaks of endonuclease activity were observed at KC1 concentrations of 0.33 and 0.45 M. The pooled fractions eluting at 0.33 M KCl, which contained about 60% of the total adsorbed activity, were dialyzed against 2 liters of 50 mv potassium phosphate (pH 6.8), 10 trm 2. mercaptoethanol, and 15% (v/v) ethylene glycol for 4 h (Fraction IV: phosphocellulose). The activity which eluted at 0.45 M KC1 was found to be specific for single-stranded DNA and will not be discussed further.
TABLE I
Purification of an endonuckzsr from cultured human KB ~11s
The details of the purification procedure for the KB cell endonu-
clease and the definition of a unit are given under “Experimental
Procedures.” The activity was measured using Assay I except for the values in parentheses where activity was measured using Assay 2.
Fraction Volume Units/ml Protein Specific activity
Effkt of diffircwt rcagcvzts on actiuity of purified c~ndonuc~lrasc
Assay 1 was used as described under “Experimental Procedures”
with the additions as indicated.
Addition
-- Number of Breaks per 3 fig Ad2 DNA
x 10 1’ Inhibition
Hydroxylapatite Chromatography - Fraction IV was loaded onto a hydroxylapatite (Bio-Rad) column (1.0 x 5 cm) that had been equilibrated with 50 nm potassium phosphate (pH 6.8), 10 nm 2- mercaptoethanol, and 10% (v/v) ethylene glycol (HA buffer). The column was washed with 20 ml of HA buffer and eluted with 30 ml of a linear gradient ranging from 0.05 to 0.5 M potassium phosphate (pH 6.81 in 10 rn~ 2-mercaptoethanol and 10% (v/v) ethylene glycol. The flow rate was 6 ml/h and l-ml fractions were collected. Endonu- clease was eluted between 0.18 and 0.24 M potassium phosphate. The peak fractions were pooled and dialyzed against 500 ml of 50 rn~ Tris-HCl (pH 7.81, 0.1 M KCl, 1 rn~ dithiothreitol, 50% (v/v) ethylene glycol for 3 h (Fraction V: hydroxylapatitel. Fraction V was stored at -80°C after the addition of bovine serum albumin to 1 mgiml and was stable for at least 3 months.
RESULTS
Purification of Endonuclease - The isolation and purifica- tion of an endonuclease from human KB cells is summarized in Table I. The maximum overall purification achieved was 750-fold with a yield of 6%.
Unless otherwise indicated, the following studies were per- formed with Fraction V endonuclease using Assay 1.
Properties of the Reaction-The pH optimum was 8.8 in 10 mM Tris-HCl buffer and glycine-KOH (pH 9.1) showed 98% activity of the maximum achieved with Tris buffer. Potassium phosphate buffer was much less effective than Tris buffer. The enzyme required divalent cations for activity. The maximal activity with MgCl, was 2 to 4 mM; maximal activity with MnCl, (2 to 10 mM) was about 50% of that achieved with MgCl,. The reaction was very sensitive to monovalent salts and was inhibited 92 and 100% by 20 and 40 mM KCl, respectively; a similar effect was observed with NaCl.
9%
None 8.2 0
tRNA (E. coli) 5 Kk! 1.2 85
10 CLg 0.2 98
20 P-23 10.05 100
p-Hydroxymercuribenzoate
0.9 rnM -co.05 100
N-Ethylmaleimide 0.9 rnM co.05 100
Actinomycin D 0.1 P.g 8.3 0
0.5 Pg 8.1 1
Ethidium bromide 0.1 M 8.3 0
6.5 PLg 8.1 1
MnCl,
4rnM 8.2 0 8mM 8.1 1
CaCl, 4rnM 4.1 50
8m~ 2.7 67
reagents (Table II). Actinomycin D and ethidium bromide, reagents which intercalate with DNA, did not inhibit endo-
nuclease activity at 8 x 10 ‘; and 2.5 x 10 ’ M, respectively. Addition of Cap+, but not Mn”+, inhibited the Mg”‘-catalyzed reaction.
Estimation of Molecular Weight by Zonal Sedimentation - The molecular weight of the purified endonuclease was esti- mated by sedimentation in glycerol density gradients. Rela- tive to standards of alkaline phosphatase, alcohol dehydrogen- ase, and bovine serum albumin, the endonuclease had a sedimentation coefficient of 3.8 S, corresponding to a molecu- lar weight of approximately 54,000 based on the assumption
Effects of Different Reagents on Endonuclease Activity- Endonuclease activity was inhibited 98% by the addition of 10 pg of E. coli tRNA (final concentration, 8 WM) (Table II). The phenomenon of tRNA inhibition of KB cell endonuclease is similar to that reported for endonuclease I from E, coli and Proteus mirabilis (16, 17). However, inhibition of T4 endonu- clease requires tRNA concentrations exceeding 1 mM (18). Although a sulfhydryl reagent was not required for activity, the enzyme was almost completely inhibited by -SH-reacting that this is a globular protein.
Substrate Specificity, Single-stranded versus Double- stranded-The enzyme degrades native Ad2 DNA but does not attack KB cell ribosomal RNA. In this respect, this enzyme is a double strand scission endodeoxyribonuclease and differs from both the endonuclease found in vaccinia virus which is specific for single-stranded DNA (19) and the endo-
FIG. 1. Agarose gel electrophoresis of the Ad2 DNA digestion products by purified endonuclease. Multiple reactions were carried out by Assay 1 with 1.0 unit of endonuclease. At various times, EDTA was added to 10 mM. The samples were loaded onto 1.4% (w/ v) agarose gels (12 x 0.6 cm) containing ethidium bromide (0.5 pg/ ml) and electrophoresis was carried out at 5 mA/gel for 3 h (20). Gels were examined by direct .illumination with UV light and photo- graphed using Polaroid type 55 P/N film (upper figure). Sample numbers refer to reactions for (1) 0 min, (2) 5 min, (31 10 min, (4) 20
nuclease reported in polyoma virus-transformed hamster cells (6) which hydrolyzes RNA.
Digestion of Ad2 DNA by the purified endonuclease was followed by electrophoresis on agarose gels and the DNA products were visualized under UV light in gels containing ethidium bromide (Fig. 11. After short digestion times, the products electrophoresed as a relatively sharp band in the agarose gel but subsequently gave rise to smaller fragments and more diffuse bands after longer digestion times. After 60 min of hydrolysis (Gel 61, the products showed considerable heterogeneity in chain length.
The distance migrated was plotted against the logarithm of molecular weight for EcoRI restriction enzyme fragments of Ad2 DNA (Fig. 2A), and the molecular weights of the digested products were estimated. The average molecular weights of the fragments of 5-, lo-, 20-, and 40-min digestions were 11.5,
I I I 1 I I 0 10 20
Reaction Time I min )
FIG. 2. A, size estimation of the digestion products of Ad2 DNA with endonuclease. Electrophoretic mobility is plotted against the log of the molecular weights of marker fragments (see Fig. 1). The mobilities of the products of (a) 5, (b) lo-, (cl 20-, and (d) 40-min digestions are indicated by the arrows. Molecular weights of Ad2 DNA and Fragments A through F of Ad2 DNA digested with EcoRI restriction endonuclease were 22.9, 13.6, 2.79, 2.34, 1.69, 1.40, and 1.12 x 106, respectively (21). B, time-dependent fragmentation of Ad2 DNA by endonuclease. Mole’cular weight of DNA fragment is plotted against the time of the reaction. a, b, and c are the products of 5-, lo-, and 20-min digestions of Ad2 DNA with endonuclease, respectively.
min, (5) 40 min, and (6) 60 min, respectively; Gel 7 is the six Ad2 EcoRI restriction endonuclease products (A throughF1 of Ad2 DNA. The lower figure is the scanning pattern of the film negative at 540 nm using a Gilford spectrophotometer with a linear transport. A through F indicate the positions of EcoRI restriction enzyme frag- ments of Ad2 DNA. The gels were also sliced into 5-mm segments, melted in 1 ml of boiling H,O, and counted in Aquasol. The results were essentially the same as displayed in the above scanning patterns.
FIG. 3. Neutral (A) and alkaline (B) sucrose density gradient centrifugation of the endonuclease oroduct. The reaction was carried out by Assay 1 with 2.3 units of purified KB cell endonuclease acting on Ad2 13HlDNA for 60 min at 37°C. The samples were centrifuged and analyzed as described under “Experimental Procedures.” Ad2 l14ClDNA provided by Dr. T. Yamashita was added as marker.
6.2, 3.8, and 0.8 x 106, respectively. These values coincided well with those (11.5, 6.5, 2.9, and 1.2 x 106) obtained from the sedimentation in neutral sucrose density gradients in a par- allel experiment. From these results, during the very early stages of the reaction, a stepwise fragmentation of the genome (23 x 10” daltons) can be demonstrated in which it is first cleaved in half, then these products are again cleaved in half, etc.
The data described above indicate that the enzyme makes double-stranded breaks on native Ad2 DNA. However, endo- nucleases which degrade native DNA via double-stranded scissions have been reported also to exhibit nicking activity on native DNA. E. coli endonuclease I which is known to degrade native DNA via double-stranded scissions also de- grades DNA via single-stranded scissions at longer digestion times or when the enzyme makes a complex with tRNA (22, 23). Native DNA is also degraded via both double- and single- stranded scissions by acid deoxyribonucleases from various sources and the single-stranded degradation becomes evident only after a time lag (24). A recent report on a T, phage- induced endonuclease has indicated that the enzyme makes single- and double-stranded breaks on native duplex DNA (18). Thus, the single-stranded scission activity of the KB cell endonuclease was examined. Ad2 [3H]DNA was treated for 60 min with purified endonuclease and the products were ana- lyzed by rate zonal sedimentation in neutral and alkaline sucrose density gradients (Fig. 3). The average molecular weights of the products in peak fractions in neutral and alkaline gradients were calculated to be 1.06 x 10fi and 0.27 x
lo6 (0.54 x 10” as double-stranded DNA), respectively (10, 25). This indicates that on the average, the enzyme makes two
single-stranded breaks/one double-stranded break after 60 min of incubation. The enzyme appears to be free of contami- nating exonuclease activity since no radioactivity was found at the top of the gradients.
The purified endonuclease has an interesting specificity for the base moiety of synthetic polynucleotides. Poly(dG) poly(dC) was hydrolyzed extensively by the enzyme, but poly(dT).poly(dA) was resistant (Fig. 4). The specificity for poly(dG) poly(dC) is reminiscent of an endonuclease activity reported to be present in highly purified penton (a structural component of the virus particle) isolated from AdB-infected cells and suggested to cleave viral DNA at GC-rich regions (28). However, the penton-associated enzyme differed in many respects from the endonuclease reported here (see “Discus- sion”).
The size of the products resulting from extensive digestion of Ad2 DNA with endonuclease was analyzed by DEAE- Sephadex chromatography and nucleic acid homochromatog- raphy. When the products of a 90-min digest were chromato- graphed on DEAE-Sephadex in the presence of 7 M urea, the formation of small oligonucleotides was not observed (Fig. 5A). Even after further digestion with fresh enzyme, signifi- cant formation of small oligonucleotides was not detected, and the product eluted as a sharp peak at a NaCl concentration of 0.56 M (Fig. 5B). This limit digest was analyzed by one- dimensional homochromatography using Homomix I (29) as a developing solvent (Fig. 6). In this system, polynucleotides with a chain length of less than 36 moved away from the origin. However, 57% of the products after the endonuclease digestion remained at the origin, indicating more than half of the products of a limit digest are longer than 36 nucleotides.
In the case of poly(dG) . poly(dC), the formation of short oligonucleotides would be expected, especially in view of the high specificity for this synthetic substrate. Contrary to this expectation, the limit product of the digestion of poly(dG) . poly(dC) contained oligonucleotides with a rela- tively long chain length, and the average chain length of the products was close to 12 to 18 as judged by DEAE-Sephadex chromatography (Fig. X).
Analysis of Bond Cleavage - The enzyme appears to form products having 3’-hydroxyl termini since the digested product is a good substrate for E. coli DNA polymerase I, as revealed by the use of Assay 2 during the purification procedure.
The products were not a substrate for polynucleotide kinase and dephosphorylation of the product with E. coli alkaline phosphatase was essential for the kinase reaction as described under “Experimental Procedures.” These results indicate the presence of 5’-phosphoryl termini in the endonuclease-di- gested products.
The degradation product of Ad2 DNA (approximately 450- nucleotide length as determined by sedimentation in an alka- line sucrose gradient) was incubated with the single strand- specific nuclease, Sl, from Aspergillus oryzae (30). During the preparation of this material, no acid-soluble radioactivity was formed. This indicates again the absence of exonuclease con- tamination in the purified endonuclease fraction (see the text for Fig. 3). Approximately 2.5% of radioactivity was rendered acid-soluble by Sl nuclease (Fig. 7). Under the conditions used, heat-denatured Ad2 DNA was degraded up to 95%, while native Ad2 DNA was hydrolyzed less than 0.3%. If the thymidylic acid residue (the labeled component) is uniformly distributed in Ad2 DNA, this result indicates that the product has single-stranded regions with an equivalent chain length of about 9 or 10 nucleotides. The single-stranded regions were
(Miles Laboratories, hybrid prepared as described previously (26)) instead of Ad2 DNA were incubated at 37°C for 0, 20, and 60 min with 11.4 units of endonuclease. Sample numbers in the figure refer to the following reactions: enzyme incubated with poly(dA). poly(dT) for (I) 0 min, (2) 20 min, (3) 60 min, and with poly(dG).poly(dC) for (4) 0 min, (5) 20 min, and (6) 60 min. The reaction was stopped by addition of EDTA to 10 mM. Electrophoresis was performed in 7% polyacrylamide gels (10 x 0.6 cm) essentially as described (27) (7 M urea was deleted) at 5 mA/gel for 2 h. Gels were stained with 2% acridine orange in 15% acetic acid overnight and destained by soaking in repeated changes of 7% acetic acid (upper figure). Gels were photographed using Polaroid type 55 P/N film (upper figure) and scanned with a Joyce-Loebl microdensito- .* - .
FIG. 4. Polyacrylamide gel electrophoresis of poly(dT1 . poly(dA) and poly(dG).poly(dC) after digestion with purified endonuclease. Six scaled up reaction mixtures (500 ~1 instead of 50 ~1) (Assay 1) containing 50 pg of either poly(dA) . poly(dT) or poly(dG1 . poly(dC1 meter (lower fzggure).
Fractm Number
FIG. 5. DEAE-Sephadex chromatography of the endonuclease di- gests of Ad2 DNA and poly(dG).poly(dC). Two lo-fold scaled up reaction mixtures of Assay 1 were incubated at 37°C with 15.2 units of the endonuclease. After 90 min, one sample (A) was heated to 100°C for 5 min after the addition of EDTA (10 mM) to stop the reaction. A lo-fold reaction mixture of Assay 1 (without Ad2 DNA) containing 15.2 units of endonuclease was added to the second mixture (B) which was further incubated for 90 min, and the enzyme inactivated as described above. Each sample was mixed with markers, 0.1 &i of 132P]dTTP and 2.5 absorbance units of (dT),,-,, (Miles Laboratories), diluted with 5 ml of column buffer (50 mM Tris-HCl, pH 7.8, 0.1 M NaCl, and 7 M urea), and loaded onto a column (0.6 x 60 cm) of DEAE-Sephadex (A-25) pre-equilibrated with the same buffer. The column was washed with 10 ml of buffer, eluted with 240 ml of a linear gradient ranging from 0.1 to 0.85 M
NaCl in the column buffer, and then with 1.5 M NaCl in column buffer. The flow rate was 10 ml/h and 2.4-ml fractions were collected. Aliquots (0.5 ml) were counted in Aquasol. The arrows indicate the positions of (a) [32P1d’ITP, and (b) (dT),,-,, (monitored by measuring absorbance at 260 nm). C, 40-fold scaled up reaction mixture of Assay 1 was prepared with 165 yg of poly(dG) poly(dC) instead of Ad2 DNA. Incubation was performed with 61 units of the endonucle- ase at 37°C for 90 min. A fresh, equal volume of reaction mixture was added (without poly(dG).poly(dC)), and the mixture was incu- bated for an additional 90 min. The sample was loaded onto a DEAE- Sephadex (A-25) column (0.3 x 60 cm) pre-equilibrated with 50 mM Tris-HCl (pH 7.81, 7 M urea, and eluted with 80 ml of a linear gradient ranging from 0 to 1 M NaCl in the same buffer as described above. The arrow indicates the position of 13HldGTP.
FIG. 6. Homochromatography of the DNA products of the endo- nuclease reaction. One milliliter from tube 70 in Fig. 5B was diluted with 5 ml of H,O and applied to a DEAE-Sephadex (A-25) column (0.5 x 2 cm) pre-equilibrated with 0.1 M triethylamine bicarbonate (pH 8.0). The column was washed with 5 ml of H,O, 5 ml of 0.1 M
triethylamine bicarbonate and eluted with 2 ml of 1 M triethylamine bicarbonate. The initial 0.5-ml flow through was discarded and the following l-ml eluate, containing all the radioactivity, was collected. The sample was dried in LKXCUO at 3o”C, dissolved in 0.1 ml of H,O, and applied onto a DEAE-cellulose plate (10 x 20 cm). Homochro- matography was carried out at 60°C as described previously using Homomix I (29). The plate was dried at room temperature and 0.5- cm wide strips of DEAE-cellulose were scrapped from the glass plate into small plastic tips, washed with 2 ml of 75% ethanol and 2 ml of 50% ethanol, and eluted with 0.2 ml of 1 M triethylamine bicarbon- ate. Each eluate was counted in Aquasol. Arrows a and b correspond to the position of the oligomers having chain lengths of 15 to 20 and 10, respectively; Arrow c corresponds to the orange dye (correspond- ing to pentanucleotide), and Arrow d corresponds to blue dye (corresponding to dinucleotide).
I I
Time I min I
FIG. 7. Extent of digestion of the endonuclease product with the single strand-specific nuclease, Sl. The Ad2 DNA digestion product (1 ml) was prepared with 3.8 units of endonuclease as described in the legend to Fig. 3. The peak of the product analyzed by sucrose gradient centrifugation corresponded to a polynucleotide of average chain length 450. The reaction mixture was extracted twice with an equal volume of phenol saturated with 30 rn~ Tris-HCl (pH 8.0). The aqueous phase was extracted with ether and dialyzed against 2 liters of 50 rn~ NaCl for 12 h at 4°C. The dialyzed DNA was digested at 37°C with Sl nuclease (30) (kindly provided by Dr. H. Natori) in 1 ml containing 33 rnM sodium acetate (pH 4.61, 1 m&i ZnSO,, 50 IXIM NaCl, and 6 pg of heat-denatured calf thymus DNA. At the times indicated in the figure, 0.1.ml aliquots were removed and 15 ~1 of calf thymus DNA (2 mgiml) and 0.3 ml of 8% cold trichloroacetic acid were added. After incubating at 0°C for 15 min, the precipitate was removed by centrifugation at 10,000 x g for 30 min and the radioactivity in 200 ~1 of supernatant was counted in Aquasol. Parallel control experiments were run with native and heat-dena- tured Ad2 DNA in which the endonuclease reaction was carried out in the presence of 20 rn~ EDTA. S, nuclease digestion of the endonuclease product (O---O), native Ad2 DNA (O- - -01, and heat-denatured Ad2 DNA (0-O).
TABLE III
End group analysis of the Ad2 DNA cwdonuclease product
The conditions for the digestion with the endonuclease, 5’ and 3’ labeling of the products, their complete digestion into mononucleo- tides, and their analysis are as described under “Experimental
Procedures.” Experiments were done in duplicate and identical
results regarding the specificity of cleavage were obtained.
Experiment Products after complete enzy- matic digestion
cPm %
5’ End analysis PdA 303 12
PdG 451 18
PdT 1,410 56
PdC 330 13
3’ End analysis d-G 12,292 14
dGp 35,365 41
dTp 19,047 22
dcp 19,918 23
not formed by exonuclease, as the contamination of exonucle- ase was ruled out as described above.
Terminal Nucleotide Determinations-As shown in Table III, the highly purified endonuclease showed considerable selectivity in its choice of cleavage sites. Analysis of the products after a 30-min digestion showed slightly more than half of the 5’ termini to be dT and somewhat less than half of the 3’ termini were dG. The fact that all four bases were represented at each end indicates that cleavage was appar- ently not restricted to specific sites, but that certain sites or regions within the DNA were preferred for cleavage by the endonuclease.
DISCUSSION
An interesting aspect of the present investigation is the substrate specificity and the mode of cleavage of the KB cell double strand scission DNA endonuclease. The enzyme shows good activity with synthetic, double-stranded polynucleotide, poly(dG).poly(dC) as substrate but is virtually inactive with poly(dT) .poly(dA). A limit digest of poly(dG) -poly(dC), con- tains products with a relatively long chain length, and the average chain length of the products was close to 12 to 18. The cleavage of Ad2 DNA showed a similar pattern. During the early stages of digestion of the double-stranded Ad2 DNA, degradation occurred such that the 23 million-dalton Ad2 genome was cleaved approximately in half (11.5 x 10” daltons) after 5 min; these were then cleaved in half (to about 6.5 x 10” daltons) after 10 min and again in half (to about 2.9 x 10” daltons) after 20 min of digestion. More than half of the final limit digest product was longer than 35 to 40 nucleotides (Fig. 6) and eluted as a sharp peak from DEAE-Sephadex in 7 M
urea (Fig. 5). Thus, in addition to specificity for nucleotide composition of the substrate, this enzyme apparently exhibits specificity for substrate size and is unable to cleave either a synthetic polynucleotide or Ad2 DNA beyond a certain point.
End group analysis of the products of a 30-min digestion of Ad2 DNA revealed approximately half of the 5’ termini contained dT and half of the 3’ termini contained dG (Table III). Thus, also with heteropolymeric DNA substrates, the endonuclease exhibited specificity with regard to both the size of the substrate and the base composition (sequence?) of the cleavage site. The fact that dT was highly represented (but not required) at the 5’ termini, yet poly(dA).poly(dT) was resistant to attack, suggests that perhaps an extended region with a certain base composition (e.g. rich in GC) is necessary for binding and cleavage.
The endonuclease that we have studied is the predominant endonuclease activity in KB cells. Application of our purifica- tion procedure to KB cells infected with Ad2 (18 h postinfec- tion) resulted in the isolation of an endonuclease with very similar, if not identical, properties. A problem of great interest is whether or not Ad2 virions contain an endonuclease and if so whether the virion-associated endonuclease is the same as the endonuclease described in this report. Such an enzyme could play an important role in the metabolism of viral or cellular DNA during infection. Burlingham and Doerfler (31) have reported the presence of an endonuclease activity in Ad2- and AdlB-infected cells that cleaved Ad DNA to 18 S DNA fragments in vivo and in vitro. A similar activity was found when purified virions were incubated with Ad DNA (32). Highly purified penton isolated from AdB-infected cells was demonstrated to contain endonuclease activity, and antibody against highly purified penton inhibited the enzyme (28). These authors (31) concluded that the Ad2 penton base serves both as an endonuclease and as a structural subunit. Based on indirect evidence (inhibition of activity by actinomycin D and by poly(dG).poly(dC)), it was proposed that the preferred site for hydrolysis is a GC-rich region. Although the KB cell endonuclease reported here shares some properties with the penton-associated endonuclease (such as its activity with double-stranded DNA and presumed preference for poly (dG) poly(dC)), it seems to be a different enzyme from the following lines of evidence. 1) The KB enzyme has a pH optimum of 8.5 to 9.0, whereas the penton enzyme is strongly inhibited above pH 8.0; and 2) the KB enzyme is inhibited 85% by 20 mM monovalent salt, whereas the penton enzyme shows constant activity up to 200 mM monovalent salt (31).
2.
4.
6.
7. 8.
9.
10. 11.
12.
13. 14.
15.
16.
17.
18. 19.
20. Share. P. A.. Sueden. B.. and Sambrook. J. (1973) Bzoch(~n~is~rv
21. Mulder. C., Arrand, J. R.. Delius, H., Keller. W., Pettersson,
Reif et al. (33, 34) have reported the presence of an endonu- clease from both uninfected and AdB-infected KB cells which had a pH optimum of 4.0. This activity could be separated from the penton-associated endonuclease (which had a pH optimum of 7.2). The pH 4.0 endonuclease was reported to be active at temperatures up to 6o”C, similar to the penton- associated endonuclease.
22. 23. 24. 25. 26.
27.
28.
Further studies on the comparative properties of these several endonucleases and fingerprint analyses of highly pur- ified enzymes might reveal their relatedness and possibly their function during adenovirus infection.
29.
30. 31. 32.
REFERENCES 33.
1. Weld, W. S. M., Green, M., and Buttner, W. (1978) in Thr Molecular Biology ofAnimal Viruses (Nayak, D. P., ed) Vol. 34. 2. DD. 673-738. Marcel Dekker. Inc. New York
Lonberg-H&n, K., and Philipson, L. (1974) in Momgraphs iu Vimlog,y, S. Karger, Base1
Groneberg, J., Brown, D. T., and Doerfler, W. (1975) Virr1log.v 64.115-131
Churchill. J. R., Urbanczvk. J.. and Studzinski. G. P. (1973) Biochrm. Biophys. Rcs. &mmun. 33. 1009-1016
Brent, T. P. (1975)BiwhinL. Biophys. Acta 407. 191-199 Koh, J. K., Waddell, A., and Aposhian, H. V. (1970) ,/. Bzol.
Chem. 245, 4698-4707 Yamashita, T., and Green, M. (1974)J. Vi&. 14, 412-420 Landgraf-Leurs, M., and Green, M. (1971) J. Mol. Bid. 60. 18%
202 Rhoades, M., and Meselson, M. (1973) </. Riol. Chcm. 248. 521-
527 Burgi, E., and Hershey, A. D. (1963)Biophy.s. ,J. 3, 309-321 Richardson, C. C. (1965) Prw. Natl. Amxl. &I U.S.A. 31. 158-
165 Sulkowski, E., and Laskowski, M., Sr. (1971) Biwhirn. Bi&,w.
Ada 240, 443-447 Wu, R. (197O)J. Mol. Bid. 51. 501-521 Roychoudhury, R., R&her, D., and Ktjssel, H. (1971) Bioc~/wv.
Bio~hvs. Rcs. Commun 45. 430-435 1 ._
Lowry, 0. H., Rosebrough, N. J., Farr, A. L.. and Randall, R. J. (195115. Biol. Chc~m. 193. 265-275
Lehman. I. R.. Roussos. G. G.. and Pratt. E. A. (1962) J. Rio2. Chem.‘237, 819-828
Goebel. W.. and Helinski. D. R. (1971) J. Rid. Chcm. 246, 3851- 3856
Kemper, B., and Hurwitz, J. (1973)5. Biol. Chum. 248, 91-99 Rosemond-Hornbeak, H., and Moss, B. (1974) J. Bid. (‘hem.
249, 3292-3296
12:3055-3063 ”
U., Roberts, R. J., and Sharp, P. A. (1974) (,‘old Spring Hnrhr symp. Quant. Bid. 39. 397-400
Bernardi, G., and Cordonnier, C. (1965)J. Mol. Biol. II. 141-143 Goebel, W., and Helinski, D. R. (1970)Bioc.hemistr.y 9. 4793-4801 Bernardi, G., and Sadron, C. (1964)Biochrmistr:y 3. 1411-1418 Studier, F. W. (1965) CJ. Mol. Bid. 11. 373-390 Tsuruo, T., Hirayma, K., and Ukita, T. (1975)Biwhim Rzoph,v.s.
Acta 383, 274-281 Grandgenett, D. P., and Green, M. (1974) J. Bid. Chum. 249,
5148-5152 Burlingham, B. T., Doerfler, W., Pettersson, U., and Philipson,
L. (1971) J. Mol. Bid. 60, 45-64 Jay, E., Bambara, R., Padmanabhan, R., and Wu, R. (1974)
Nuc~leic Acids Rrs. 1. 331-353 Vogt, V. M. (1973) Eur. J. Biochem. 33. 192-200 Burlingham, B. T.. and Doerfler, W. (1972)J. Vird. 48. 1-13 Marusyk, R. G., Morgan, A. R., and Wadell, G. (1975) J. Vird
16, 456-458 Reif, U., Winterhoff, U., and Doerfler, W. (1977) Eur. J. Hu-
