Deoxyribonucleic acid synthesis in bacteriophage SP01-infected Bacillus subtilis. II. Purification and catalytic properties of a deoxyribonucleic acid polymerase induced after infection.

Abstract A new DNA polymerase has been purified 1000-fold from Bacillus subtilis infected with bacteriophage SP01. It consists of a single polypeptide chain of approximately 122,000 molecular weight. The enzyme requires deoxyribonucleoside 5'-triphosphates, Mg2+, β-mercaptoethanol, and single stranded DNA templates for activity. Bihelical DNA which has been partially denatured serves as the preferred template in in vitro reactions. Although 5-hydroxymethyluracil replaces thymine in SP01 DNA, deoxynucleotide analogues of 5-hydroxymethyldeoxyuridine 5'-triphosphate can serve as efficient substrates for the phage polymerase. The purified enzyme is free of endonuclease activity.


SUMMARY
A new DNA polymerase has been purified lOOO-fold from Bacillus subtilis infected with bacteriophage SPOl. It consists of a single polypeptide chain of approximately 122,000 molecular weight.
Bihelical DNA which has been partially denatured serves as the preferred template in in vitro reactions.
The purified enzyme is free of endonuclease activity.
DNA polymerases have been isolated and extensively characterized from a number of biological systems. However, semiconservative DNA replication has not been accomplished by these enzymes in vitro. Bacillus subtilis and some of its phage systems offer a novel approach in that template DNA and product can be biologically characterized by means of the transformation and transfection systems. Phage systems are less complex because of their smaller chromosome size, number of DNA polymerase activities coded by the genome, and thus provide a better opportunity to examine the mechanism of DNA replication in vitro.
Earlier studies (1) provided evidence that a new DNA polymerase activity is induced upon infection of B. subtilis by the virulent phage SPOl. Further evidence that this new polymerase activity is coded by the phage genome requires purifica-* This investigation was supported by Research Grants GM-11108 and TO1 GM 295 from the National Institutes of Health and GB 8739 from the National Science Foundation.
The first paper in this series is Ref. 1. $ Postdoctoral fellow of the American Cancer Society (PF-578). Present address, The Rockefeller University, New York, New York 10021.
$ Recipient of National Institutes of Health Research Career Development Award GM 50199. tion and characterization of the new activity aud comparison of its properties with those of the host DNA polgmerases (2). The present report describes the purification of a new DNA polymerase which occurs in cells upon infection with SF01 and which differs in many aspects from those of the host,. These studies also reveal some similarities to the properties of other enzymes induced by virulent phages of Escherichia coli.

R'lATERIliLS
Cell Growlh and Strains-Procedures for the growth of bacteria and the preparation of phage stocks are described in the first paper of this series (1). B. subtilis strain S131060 (trpC hisB pheA polA5), a DNA polymerase I-deficient mutant (3), was used in these experiments.
Aquasol is a product of New England Nuclear.
6-(p-Hydroxyphenylazo).uracil was the generous gift of Dr. Bernard Langley of the Imperial Chemical Industries, Ltd. Stock solutions in 50 m&r NaOH were reduced immediately before use by incubation at 30" for 10 min with 20 rnhf dithiothreitol (6). METHODS L>iV;l Templates jar DNA Polymerase Reaction-DNA preparations were purified as described previously (1). DNA was "activated" by limited digestion with pancreatic DNase I (7), and portions were further digested until 5 to 10% of the material became acid-soluble with E. coli exonuclease III to produce molecules containiug gaps. Denatured DNA was prepared either by boiling for 5 min and fast cooling or by alkali (0.1 N NaOH) treatment for 5 min and neutralization.
Preparation of IVztcleotide Kinase-A partially purified enzyme extract of SPOl-infected cells catalyzes the phosphorylation of hmdUXl', with the use of ATP, to its triphosphate. SB19 (3 g), collected at 15 min after infection with SPOl, was suspended in 30 ml of 0.02 M Tris, pH 7.5, 0.01 ar fi-mercaptoethanol, 0.1 MM EDTX and treated with lysozyme (200 pg per ml) at 37" for 5 min. The lysate was sonicatctl at the maximum output of a l{ranson Sonifier for 5 mm, centrifuged at low speed to remove cell debris, and followed by further centrifugation at 100,000 x g for 60 min. The protein concentration of the 100,000 x g supernatant was adjusted to IO mg per ml, and 0.2 ml of freshly prepared 5% streptomycin sulfate per ml was added slowly while the solution was stirred.
The suspension was centrifuged at 14,500 X g for 20 min. The supernatant fraction was precipitated with a saturating amount of (NH&SO+ centrifuged, and the precipitate dissolved in 0.04 &I Tris, pH 7.6, 0.1 or KCl, 0.01 11 fl-mercaptoethanol, 1 rn;\r MgCl,, and 0.1 ml{ EDTA, dialyzed against the same buffer and stored at -15". Combined nucleoside kinase and nucleoside diphosphokinase activity was measured by the conversion of [3H]dAMP to [3H]-dATP by the reaction described below.
The triphosphate was separated from unreacted monophosphate by chromatography on polyethyleneimine cellulose with 0.4 M NH,HCO, as the solvent.
Radioactivity was located by counting segments of the lane in which the material was chromatographcd.
Synthesis of 5-Hydroxymethyldeoxyuridine 5'-Triplaosplaate-SPOl DNA was enzy-matically digested with pancreatic DFase I and venom phosphodiesterase, and the resulting deoxynucleoside monophosphates were separated by paper chromat.ography. hmdUlIP, recovered from the paper, was used as a substrate for the nucleotide kinase from SPOI-infected cells to yield the corresponding deoxynucleoside triphosphate.
Xl'01 DNA (4400 nmoles) in 2.0 ml of 0.02 M Tris, pH 7.5, 0.02 M NaCl, 0.01 M MgClz was hydrolyzed with crystalline pancreatic DNase I (2 mg) for 3 hours at 37". The pH was raised to 8.6 by the addition of NH,OH, and the polydeoxyribonucleotidcs were digested to mononucleotides by treatment with venom phosphodiesterase (1 mg) at 37" for 3 hours. Conversion of monophosphates to deoxynucleosides by contaminating 5'-nucleotidase was minimal.
The digestion mixture was heated in a boiling water bath for 2 min, chilled, and the flocculent pre- and 0.8 mg of the SPOl nucleotide kinase preparation.
After 60 min of incubation at 37", the reaction mixture was heated at 80" for 2 min and centrifuged at 15,000 x g for 15 min. The supernatant was chromatographed as described above and the resulting deoxynucleoside triphosphate band eluted with 0.1 mM EDTA. hmdUTP was obtained in approximately 80 y0 yield. DNA Polymer-use Assay-The assays measured the incorporation of labeled deoxynucleoside monophosphates from triphosphate substrates into acid-insoluble material. The reaction mixture (100 pl) contained 100 mM Tris, pH 7.5; 50 rnx NaCl; 10 nlM @-mercaptoethanol, 15 m&f MgClz; 0.1 mM EDTA; 50 PM (each) of dCTP, dGTP, dATP, dTTP, one of which contained 3H label with a specific activity of 20 to 40 cpm per pmole; 75 PM DNA t.emplate; and enzyme. When poly[d(A-T)] was used as template, dCTP and dGTP were eliminated from the reaction mixture.
Incubations were at 37" for 30 min, and 1 unit of enzyme is defined as the incorporation of 10 nmoles of total nucleotide into acid-precipitable product under these conditions (7).
Endonuclease Assay-Endonuclease activity was estimated by measuring the decrease in genetic linkage of closely associated markers on the B. subtilis chromosome by transformation. The aromatic linkage group carries at least nine genes specifying the biosynthesis of aromatic amino acids and includes a histidine marker.
Sl3202, the strain used in this assay, carries four widely distributed markers of this linkage group (9). Single strand breaks in the DNA cause a loss of linkage of two markers when they occur within the DNA segment containing these markers (lo), thus providing a very sensitive assay for endonucleolytic cleavage. The efficiency of a DNA preparation from wild type cells to transform all four markers following exposure to the DNA polymerase fractions was tested by using nonsaturating concentrations of the standard DNA. SB19 DNA (1 nmole per ml) was incubated under assay conditions in t.he presence of enzyme fractions at 30". Competent cells of SB202 (1 x lo8 cells per ml) in minimal medium plus 0.5'$& glucose were added, and the mixtures were incubated further at 30" for 30 min. Samples were diluted and Trp+ recombinants were selected. il minimum of 300 Trp+ recombinants were stroked onto nutrient agar plates. These were replica plated to appropriately supplemented media to determine the remainder of the genotype.
Exonuclease Assay-Exonuclease activity in the absence of measurable endonuclease activity was assayed by the production of acid-soluble radioactivity with the use of [3H]poly-[d(A-T)] as substrate.
Incubation mixtures (100 ~1) contained 33 mM Tris, pH 8.0; 3 rnhf EDTA; 1 mM dithiothreitol; (92.1 X lo4 cpm per nmole); and enzyme. Following incubation at 37" for 30 min, the reaction mixtures were chilled, supplemented with 200 pg of salmon sperm DNA, and made 5% in trichloroacetic acid. The resulting precipitate was sedimented at 25,000 x g for 10 min, and the supernatant was counted in 10 ml of Aquasol by liquid scintillation spectrometry.
Alternatively, exonuclease activity was measured by the release of mononucleotides from P22 phage [aH]DNA. Native or denatured DNA (12 nmoles, 1.2 X lo4 cpm per nmole) was incubated with enzyme under the conditions of the DNA polymerase assay, except that the deoxynucleoside triphosphates were omitted.
Aliquots of the mixture were spotted on strips (1% x 9 inch) of DE81 paper, and mononucleotides were eluted from the origin by chromatography with 0.35 M ammonium formate (pH 8). The dried strips were cut into l-cm segments, and the radioactivity of each w-as determined by liquid scintillation counting.
The amount of exonuclease activity was determined from the proportion of radioactivity left at the origin as unhydrolyzed polynucleotide compared to the radioactivity which chromatographed as mononucleotide. Gel Electrophoresis-Polyacrylamide disc gel electrophoresis was carried out as described by Davis (11). The procedure of Weber and Osborn (12) was used for electrophoresis in the presence of SDS. Samples were heated in a boiling water bath for 2 min in the presence of 1% each of SDS and P-mercaptoetha-no1 prior to SDS gel electrophoresis.
Gels were scanned at 580 nm after staining with Coomassie blue with the use of a Gilford model 2000 recording spectrophotometer equipped with a linear transport mechanism. General-Protein concentration was determined by the method of Lowry et al. (13). DNA concentration was estimated by the method of Burton (14). Sucrose gradient analyses were performed as described by Martin and Ames (15).

Purification
Extracts of SPOl-infected B. subtilis display a significant increase in DNA polymerase activity over that observed in uninfected cell extracts (1) and which is efficiently blocked by chloramphenicol ( Fig. 1). This observation suggested that a new polymerase might be induced by the phage which required protein synthesis.
Phage-infected cell extracts were subjected to further purification in an attempt to characterize this enzyme. All procedures were carried out at 4", and all buffers contained 1 mM dithiothreitol or 10 mM P-mercaptoethanol unless otherwise indicated.
A summary of the purification is presented in Table I.
Preparation of SPOI-infected Cells-Strain SBlO60 (trpC hisB pheA polil5) was grown in 20 liters of minimal medium (16) supplemented with 0.1% casein hydrolysate and 20 pg per ml each of the required amino acids at 37" to a cell concentration of 4 x lo8 per ml. The culture was infected with SPOl at a multiplicity of 5 phages per bacterium and allowed to continue shaking for 20 min. Cells were harvested by centrifugation after the addition of NaN3 to a final concentration of 0.01 RI.
Preparation oj Cell &tract-The phage-infected cells were suspended in 0.25 nz sodium phosphate buffer, pH 6.8, to a final volume of 240 ml. The suspension was then incubated at 37" for 30 min in the presence of 50 pg per ml of egg albumin lysozyme and 20 pg per ml of ribonuclease.
The viscosity was reduced by brief treatment in a Waring Blendor.
The lysate was then brought to a final volume of 530 ml by the addition of more buffer (Fraction I).
Phase &Arc&ion-Nucleic acids were removed from Fraction I by extraction in a polyethylene glycol-dextran phase partition system (17). To Fraction I was added 120 g of 30% (w/w) polyethylene glycol and 72 g of 20 y0 (w/w) dextran T500 to give final concentrations of 5% and 2a/ respectively. The homogenate was stirred for 2 hours and then centrifuged at 500 x g for 15 min. The resulting top phase was removed and replaced by an equal volume of upper phase from a previously prepared mixture. NaCl (61.2 g) was slowly added to the solution to give a final concentration of 1.5 M, mixed, and centrifuged as above. NaCl was removed from the resulting upper phase by extensive dialysis against 0.1 M sodium phosphate buffer, pH 7.6, giving a final volume of 670 ml (Fraction II).

74.59
Ammonium Sulfate Precipitation-Solid ammonium sulfate (134 g) was slowly added to Fraction II with continuous stirring, and the mixture was allowed to stand in a separatory funnel. After 8 hours, a lower phase containing the enzyme separated from the upper polyethylene glycol phase. Further ammonium sulfate (0.3 g per ml) addition resulted in the formation of a precipitate which was removed by centrifugation at 13,260 x g for 90 min. The precipitate was dissolved in 90 ml of 0.1 M Tris, pH 7.6, containing 1 mM MgCh (Fraction III).
DEAE-Xephadex Chromatography-A column of DEAE-Sephadex (5 cm2 X 15 cm) was prepared and equilibrated with 0.1 M Tris, pH 7.6, containing 1 mM MgC12. Fraction III was diluted with buffer and absorbed to the column at a rate of 0.5 ml per min. The column was then washed with 100 ml of the above buffer and eluted with a 450-ml gradient of NaCl from 0 to 0.5 M final concentration.
The flow rate was adjusted to 0.5 ml per min, and 5-ml fractions were collected and assayed for enzyme activity.
The enzyme eluted at 0.35 M NaCl, and all of the active fractions were pooled and concentrated on an Amicon apparatus equipped with a XM-50 (50,000 molecular weight cut-off) filter.
The concentrated enzyme solution was dialyzed for 6 hours against 0.1 M potassium phosphate buffer (pH 7.6) containing 20 y0 glycerol and stored at -15" until used (Fraction IV).
Phosphocellulose Chromatography-A portion of Fraction IV (7.5 ml) was dialyzed against 0.1 M potassium phosphate buffer (pH 6.5) and applied to a phosphocellulose column (0.6 cm2 x 15 cm) previously equilibrated with the same buffer. The column was washed and a linear gradient (80 ml) of 0.1 to 0.7 M potassium phosphate buffer was applied.
Two-milliliter fractions were collected, and the DNA polymerase was eluted at a buffer concentration of 0.48 M. The fractions containing polymerase activity were pooled (Fraction V). This fraction contained 63% of the DNA polymerase activity present in the crude extract and represented a 1040-fold purification.
Fraction V was concentrated by the addition of (NH&S04 (0.5 g per ml) and dissolving the precipitate collected by centrifugation in 0.1 M potassium phosphate buffer (pH 7.6) containing 40% glycerol. The concentrated solution had a protein concentration of 1.2 mg per ml and was stable upon storage at -15" for several months.
The characteristics of the enzyme activity in Fractions IV and V studied in the following sections were similar, and several experiments that required large amounts of enzyme were performed with the use of Fraction IV as indicated.
The enzyme showed linear incorporation over a 30-min period, leveling off after 60 min. Incorporation was linear over a range of 0.04 to 0.4 unit of polymerase.
Polyacrylamide gel electrophoresis of samples of Fraction V (Fig. 2) revealed the presence of three visible bands.
The SPOl DNA polymerase band, located by assaying gel slices, comprised 60% of the total protein.
SDS gels also contained three bands and indicated that the polymerase consists of a single polypeptide chain.

Properties of XPOI DNA Polymerase
Reaction Requirements-As in the case of other reported DNA polymerases, omission of one of the deoxyribonucleoside triphosphates, MgC12, or template DNA reduces the level of incorporation to negligible amounts (Table II). Absence of /3mercaptoethanol in the reaction reduced the extent of incorporation by half, whereas lack of EDTA had no effect. When the reaction was carried out in the absence of &mercaptoethanol and in the presence of either p-hydroxymercuribenzoate, or N- Aliquots (20 pg) of Fraction V were subjected to electrophoresis in the absence (top) and presence (bottom) of O.ly', SDS as described under "Methods." The dye marker is described by the peak on the right side of the lop profile. it was inhibited to an extent of 96% and 87%,, respectively.
The Mg++ requirement could be partially replaced by MnC12. Optimal activity at a Mn* concentration of 0.3 mM decreased to less than 1 To with increasing or decreasing concentrations.
The reaction was not inhibited by the presence of 200 PM 6-(p-hydroxyphenylazo)-uracil in the assay mixture. The optimal pH range for the enzyme was 7.3 to 7.5 with either sodium or potassium phosphate and Tris buffers (Fig. 3). The enzyme was similarly active with all buffers. ma1 denaturation temperature are 2-to a-fold more efficient in synthesis than completely denatured templates.
Xolecular Weight Determinations-The molecular weight of the phage polymerase was estimated by three methods.
First, the enzyme was subjected t,o zone sedimentation on 5 to 20% linear gradients of sucrose as described by Martin and Ames (15). The sucrose solutions contained 50 InM Tris, pH 7.6, 10 mM fimercaptoethanol, and 0.1 mM EDTA. The samples were centrifuged at 35,000 rpm in the SW 39 rotor of a Spinco L2-65 preparative ultracentrifuge for 29 hours at 5". E. coli DNA polymerase I and human hemoglobin served as standards.
As shown in Fig. 4, the polymerase activity presented a unimodal distribution, sedimenting as a globular protein of 129,000 daltons. The enzyme was also chromat,ographed .on Sephadex G-200 along u-ith bovine -y-globulin and human hemoglobin.
A molecular weight of 123,000 was calculated in this case. Again, the enzyme activity eluted in an unimodal fashion.
Thirdly, the enzyme migrated as a single band during electrophoresis in the presence of 0.1% SDS (Fig. 2) with a mobility slightly greater than that of P-galactosidase.
Based on a molecular weight for /3-galactosidase of 130,000 (la), the SPOl DNA polymerase has a molecular weight of 122,000.

Deoxynucleotide
Analogue Incorporation-A number of deoxgnucleotide analogues of hmdUT1' serve as efficient substrates for the SPOl DpiA polymerase as shown in Table IV. The ability of these analogues to substitute for the natural substrate is determined by their hydrogen-bonding specificities. Addition of increasing amounts of hmdUT1' to reaction mixtures in which the 3H label is contained in dTTI' resulted in reduced incorporation. Endonuclease Activity-When SB19 DNA was exposed to 0.2 unit of SPOl DSA polymerase (Fraction V), no decrease in the subsequent assay for Trp+ transforming activity was observed. stored at -15", remained quite stable over a period of several When these recombinants were examined for the cotransfer of all months. The enzyme rapidly lost activity when heated at four markers of the linkage group using SU202, again there was temperatures above 50". Incubation of the enzyme in the no loss of linkage (Table V). In fact. the linkage analyses presence of deoxynucleoside triphosphates resulted in increased shoycd an increase compared with that of the control. stability at temperatures below 50". Exonuclease Activity-When Fraction IV was assayed for exonuclease activity on [3A]poly[d(A-T)], an appreciable level was DISCUSSION found. However, the majority of this activity failed to sediment with the phage polymerase in sucrose gradients (Fig. 4). Only a minor portion of the activity coincided with the DNA polgmerase profile.
Addition of deosynucleoside triphosphates to the reaction mixture produced no appreciable decrease in activity.
Esonuclcase activity was also found in Fraction V (Table VI).
In this case, hydrolysis was measured by DE81 paper chromatography of the reaction products. The maximal rate of hydrolysis of the substrates examined xyas obtained with denatured DNA. Enzyme StabilifU-Fraction IV, made ZOc/l in glycerol and Advantage was taken of the fact t.hat the 81'01.induced DNA polymerase was tightly bound to the I>NA-containing fractions (1). The initial step of the polyethylene glycol-destran phase extraction procedure was carried out at low salt concetltration. Under these conditions the phage polymerase remained associated with the nucleic acid fraction and partitioned in the destran phase. In the second step the salt concentration was increased and the enzyme became disassociated from the nucleic acid. In this case, the enzyme partitioned into the upper polyethylene glycol phase resulting in a A-fold purification.
Earlier evidence that a new DNA polymerase is induced upon the infection of B. su6tilis by 8POl is supported by a number of T.IRI,I,: IV observations. DNA polymerase I activity disappeared following Specijkdy of Sl'Ol polynwase for tleuxy~ibonucleosice triphosphates XI'01 infection of wild type cells (1). The purification of the The 1)NA polymernse reaction mixtures contained 10 nmoles of phage enzyme from a pal-strain suggests that DNA polymdenatllred SPOl I)NA as template, dGTP, dATP, [3H]dCTP, and erase 1 of the host is not involved in phage DKA replication. 0.08 unit of Fraction IV. The deoxyribonrlcleoside triphosphstes Chromatography of the enzyme on DEAE-Sephades clistinlisted were added and assayed as described lmder "Methods." auishes it from DNA Dolvmerase II purified from S1%1060 (2). differ greatly in size, stability, and cellular concentration.

_ "
Both ~ I DNA polymcrases II and III, having molecular weights of 140,000 to 150,000 and 160,000 to 180,000 (a), respectively, are DNA polymerase III from B. subfilis and the SPOl enzyme do larger than the S1'01 polymcrase and are much less stable upon chromatograph similarly on anion exchange columus but they storage than the phage rllzyme.2 SPOl polymerasc is also distinguished from host DX.1 polymerasc III by nature of the 6-(p-hydroxyphenylazo)-uracil sensitivity of the latter (18) compared to the resistance of the phage enzymc.3 Roth of the enzymes from SI31060 are more heat-sensitive than the phage polymerase.
DNA polymerases II and III of the host and the phage enzyme are all inhibited to various extents in the presence of sulfhgdryl-blocking compounds, indicating the involvement in enzymatic activity of a sulfhydryl group in each of these proteins. Finally, the template specificity of the S1'01 polymerase differs greatly from those of the host enz?-mes.
Like the DNA polymerases induced 1 y other lytic phages (1 g-22), the phage enzyme shows a r:rcference for denatured I)iYA over native, activated, and esonuclease III-treated DNA templates. All of the host enzymes function best on double stranded templates which have been activated by pancreatic DNase and partially digested by exonuclease III.
Thus, it is unlikely that the phage DNA polymcrase activity results from the modification of any of the existing host DiYA polymerases but that a new enzyme is coded for by the phage genome.
Our observation that partially denatured DNA serves as a better template for SPOl DSh polymerasc than completely denatured DNA is similar to results obtained by Orr  show that the initial rate of synthesis on such templates is at least a-fold greater than with completely denatured templates.
The greatest proportion of exonucleasc activity in partially purified preparations of I)n'A polymerase from SI'Ol-infected cells did llot further fractionate with the phage polymerase. The majority of the esonuclease activity was separated from Dx.1 polymerase upon sucrose gradient or phosphocellulose fractiollation.
However, a small amount was present in Fraction V. 111 comparison with the exonuclease activity associated with 1'4 DNA polymerasc (20,23), the ratio of polymerase activity to nuclease activity, calculated from the molar incorporation of tleoxynucleotides compared to acid solubilization of deosynuclcotitles from labeled substrates, for the Sl'Ol DNA polymerase is relatively high. It is not possible at this stage to corrclutie that the esonuclease activity is associated with the polymcrase molecule.
R'e previously u'erc unable to detect a nuclease activity in cell-free estracts of SPOl-infected cells which would degrade poly[d(A-T)] (1). However, upon purification an esonuclease activity has been found which will cause a solubilization of counts from [3H]poly[d(A-T)].
The level of esonuclease activity is not high enough to interfere with the polymerase assay when poly[d(A-T)] was used as the template. The nuclease activity was only sufficient to degrade approxirnately 10% of the template.
Sin~~lc strand breaks in a duplex DNA molecule such as those h introduced by endonucleolytic attack are capable of causing a large decrease in the cotransfer of linked markers in a transformation assay (10). Incubation of SPOl DNA polymerase from Fraction V with transforming DNA of B. subtilis led to no decrease in the linkage of markers of the aromatic amino acid group.
Similar results were observed (24) in assays for loss of transforming activity upon exposure of B. subtilis I)NA to the most purified fraction of E. co2i DNA polymerasc I. The linkage analyses performed here provide a 6-to 10.fold increase in sensitivity of this assay compared to that for single markers. Treatment of transforming DNA with 2 x 10e3 pg per ml of DBase I at 0" for as little as 2.5 min is capable of reducing the linkage of these markers by 80% (10). Our failure to detect my endonuclease activity in the polymerase peak fractions does not totall\-eliminate the possibility that such an activity exists, but the extent of any activity associated with the phage polymerase would be extremely low.
The catalytic properties of the SPOl-induced polymerase are similar to those of the enzymes induced by T2 (19), T4 (20), T5 (21), and 1'7 (22). The requirement for a single stranded DNA ternplate and failure to efficiently use "activated" DNA is common to these enzymes. 111 contrast to the other phage polymerases, the 81'01 enzyrne functions optimally with its own denatured l>NA template compared with salmon sperm or B. subfilis DNA denatured in a similar manner.
This might be reflected by the state of collapse of the templates as discussed. Although genetic evidence was presented that these new polymerases are required for phage DNA replication, their role in replication of bihelical DKA in vivo is not understood.
In addition to the known requirements for polymerase, ligase, and nuclenses, it is likely that other enzymes and factors are necessary to initiate and sustain sequential replication.
The gene 32 protein of T4 which binds to T4 DNA was found to stimulate synthesis n ith T4 single stranded DNA template (25). It was recently shown that RNA can initiate conversion of the single stranded DNA of Ml3 (26) or 4X174 (27) to its replicative form. In addition, +X174 conversion requires the B. coli dnarl function for initiation and the dnaU function continuously. A similar reaction could be postulated to explain the mechanism by which phage DNA polymerases initiate DNA replication on a double stranded template in vivo. Accordingly, the polymerases could function at locally denatured regions by a similar mechanism and replicate discontinuously as shown by Sugimoto et al. (28).
1n vivo replication in many phage systems involves the formation of DS.1 concatemers.
To account for the formation of concatemcrs and Okazaki pieces, Watson has proposed a modified knife and fork model (29) based on recent studies of T7 DNA replication (30). SPOl was shown to form replication intermediates with molecular weights in excess of that of DNA from mature phage particles (31). It is thus likely that SPOl DNA replication involves the formation of concatemers and might follow a course similar to that of T7.