Gene 4 Protein of Bacteriophage T7 PURIFICATION, PHYSICAL PROPERTIES, AND STIMULATION OF T7 DNA POLYMERASE DURING THE ELONGATION OF POLYNUCLEOTIDE CHAINS*

With the use of an in vitro complementation assay to measure activity, the gene 4 protein of bacteriophage T7 has been purified 1000-fold to yield a nearly homogeneous protein. The purified gene 4 protein is a single polypeptide having a molecular weight of 58,000. In addition to being essential for T7 DNA replication in vivo and in vitro, the gene 4 protein is required for DNA synthesis by the purified T7 DNA polymerase on duplex T7 DNA templates. In the absence of ribonucleoside 5'-triphosphates, DNA synthesis by the gene 4 protein and the T7 DNA polymerase is dependent on phosphodiester bond interruptions containing 3'-hydroxyl groups (nicks) in the duplex DNA. The reaction is specific for the T7 DNA polymerase, but any duplex DNA containing nicks can serve as template. The Km for nicks in the reaction is 3 x 10(-10) M.

With the use of an in vitro complementation assay to measure activity, the gene 4 protein of bacteriophage T7 has been purified lOOO-fold to yield a nearly homogeneous protein. The purified gene 4 protein is a single polypeptide having a molecular weight of 58,000. In addition to being essential for T7 DNA replication in uiuo and in vitro, the gene 4 protein is required for DNA synthesis by the purified T7 DNA polymerase on duplex T7 DNA templates. In the absence of ribonucleoside 5'-triphosphates, DNA synthesis by the gene 4 protein and the T7 DNA polymerase is dependent on phosphodiester bond interruptions containing 3'-hydroxyl groups (nicks) in the duplex DNA. The reaction is specific for the T7 DNA polymerase, but any duplex DNA containing nicks can serve as template. The K,,, for nicks in the reaction is 3 X lo-lo M.
Genetic analysis of bacteriophage T7 (l-3) has shown that the product of gene 4 of the phage is essential for the replication of the phage DNA. In cells infected with T7 phage containing an amber mutation in gene 4, only a small amount of phage DNA is synthesized. Studies with cell-free systems (4,5) have demonstrated that the product of gene 4 is also required for T7 DNA replication in vitro. This in vitro requirement for the gene 4 protein has been the basis of assays for the gene 4 protein that have led to its partial purification (6-9).
The purified gene 4 protein effects a marked stimulation of DNA synthesis by the T7 DNA polymerase on duplex DNA templates (7,9,10); T7 DNA polymerase alone is unable to use such templates (11,12). Our early studies (7)  implicating the gene 4 protein in the initiation of polynucleotide strands. Recent studies indicate that the stimulation of DNA synthesis by the gene 4 protein on duplex DNAs may be accounted for by two novel activities of the protein: an oligoribonucleotide-synthesizing activity (9);' and a DNA-dependent nucleoside 5'-triphosphatase activity (13). The former activity is involved in initiation of polynucleotide chains by the T7 DNA polymerase, while the latter may facilitate unwinding of duplex DNA during elongation of polynucleotide chains.
In uiuo studies on mutants of phage T7 suggested a role of the gene 4 protein in the initiation of DNA synthesis. Wolfson and Dressler (14) found that, after shifting a gene 4 temperature-sensitive mutant to the nonpermissive temperature, large single-stranded gaps were formed on one side of each growing fork, suggesting that the gene 4 protein functions in the initiation of DNA synthesis on the lagging parental strand. Consistent with this interpretation is Stratling and Knippers' observation (6) that the DNA synthesized in the absence of a functioning gene 4 protein hybridized to the H strand exclusively. In vitro, significant amounts of DNA synthesized on T7 DNA templates by T7 DNA polymerase . and gene 4 protein are not covalently attached to the template molecules (7,9). Scherzinger and Litfin (15) showed that extracts of T7-infected Escherichia coli require the gene 4 protein for conversion of single-stranded circular phage DNAs to circular duplexes. Recently, Scherzinger et al. (9) Ll (18), and incubated at 43". The turbid plaques were picked, grown on E. coli 011' and the resulting phages were tested for the presence of amber mutations in genes 3, 5, and 6 by complementation against phages containing individual amber mutations (3). Approximately 10% of the plaques contained T73,5,w,.

Purification of NTP
The dNTPs were treated with NaI04 to destroy rNTPs and were then chromatographed on DEAE-Sephadex A25 as described by Wu (28 (31), and T4 polynucleotide ligase (26) were assayed as previously described.
All of the above enzymes were diluted into 10 rnM Tris (pH 7.51, 10 mM Z-mercaptoethanol, 0.5 mg/ml of bovine serum albumin. Pancreatic DNase was assayed according to Kunitz (32). Alcohol dehydrogenase was assayed according to Vallee and Hoch (33 Electrophoresis in the presence of 0.1% sodium dodecyl sulfate was oerformed according to Weber and Osborn (37). Electroohoretic analysis under native conditions was performed in the Tris system of Jovin et al. (38). Gels were stained with Coomassie brilliant blue (39), and the protein was quantitated by scanning with a Joyce Loebl recording microdensitometer.

Other Methods
Protein was determined by the method of Lowry (40) using bovine serum albumin as a standard. All pH measurements were made at room temperature at a buffer concentration of 0.05 M.

Purification of T7 Gene 4 Protein
An earlier procedure for the purification of the gene 4 protein from T&,,,,;-infected Escherichia coli resulted in a preparation that was approximately 25% pure as judged by polyacrylamide gel analysis (7). A major contaminant was the T7 DNA ligase that represented over 50% of the protein in the most highly purified fraction. For this reason we constructed the T7s,5.6.dig phage carrying a deletion of gene 1.3, the ligase gene (see "Experimental Procedures"). The use of this phage for the preparation of the infected cells eliminates T7 DNA polymerase and DNA ligase activities during purification.
The following procedure, a modification of that described by Hinkle and Richardson (7), has been used to purify the T7 gene 4 protein from T7s,5,,,di,-infected E. coli DllO. A summary of a typical purification from 400 g of infected cells is presented in Table I. Unless indicated otherwise, all steps were performed at 4", and centrifugation was at 8000 rpm for 30 min in a Sorvall GS3 rotor.

Growth
of Phage-infected Cells-E.
coli DllO was grown and infected with T7:s,5,,i,dis as previously described (7). The cell paste was suspended in 0.05 M Tris (pH 7.5)/10% sucrose at a final volume of 4 ml/g of cells. Aliquots of the suspended cells (67 ml) were distributed into Spinco 45-Ti rotor tubes, frozen in liquid nitrogen, and stored at -85".

Preparation
of Cell Extract-Twenty-four tubes of suspended frozen cells (67 ml/tube) from 400 g of cells were thawed overnight on ice, and 1.5 ml each of 5 M NaCl and 10 mg/ml of lysozyme, 50 mM Tris (pH 7.51, 10% sucrose were added to each tube. After 45 min at O", the tubes were transferred to a 37" water bath, heated to 20" with stirring, and then chilled to 5" in an ice bath with stirring. The lysates were then centrifuged at 40,000 r-pm for 45 min in a Spinco 45-Ti rotor. The supernatant fluids were decanted and saved; the loose upper layer of the pellets was removed, added to 200 ml of 10% sucrose, 100 mM NaCl, 50 mM Tris (pH 7.51, and the mixture was centrifuged at 10,000 r-pm for 30 min in a Sorvall GSA rotor. The supernatant fluids were pooled and adjusted to. give an AzeO of 200 by the addition of 10% sucrose, 100 mM NaCl, 50 mM Tris (pH 7.5), yielding a final volume of 1508 ml (Fraction I). Streptomycin and Ammonium Sulfate Fractionation-To 1508 ml of Fraction I were added 151 ml of a freshly prepared solution of 30% (w/v) streptomycin sulfate over a 30-min period with stirring. After stirring for an additional 30 min, the precipitate was removed by centrifugation (Fraction II). Ammonium sulfate (499 g) was then added to the supernatant fluid (1595 ml) with stirring over a 30-min period. After stirring for an additional 30 min, the precipitate was collected by centrifugation and was dissolved in 1500 ml of 20 mM Tris (pH 7.51, 0.1 mM EDTA, 10 mM 2-mercaptoethanol, 10% (v/v) glycerol (Buffer A) (Fraction III).

DEAE-Cellulose
Chromatography-A column of Whatman DE52 DEAE-cellulose (62 cm2 x 33 cm) was prepared and washed with 12 liters of Buffer A containing 0.1 M NaCl. The concentration of (NH&SO, in Fraction III was determined by measuring its conductivity. The fraction was diluted to 4 liters with Buffer A to reduce the (NH&SO, concentration to 30 mM and then applied to the column at a rate of 900 ml/h. The column was washed with 2 liters of Buffer A containing 0.1 M NaCl, and the proteins were eluted with 16 liters of a linear gradient from 0.1 to 0.4 M NaCl in Buffer A applied at 900 ml/h. Gene 4 complementing activity eluted at approximately 0.24 M NaCl. Fractions having a specific activity of 80 units/mg or greater were pooled (1850 ml), and the protein was precipitated with (NH&SO, (390 g/liter). The precipitate was collected by centrifugation, dissolved in 40 ml of 20 mM potassium phosphate buffer (pH 7.01, 0.1 mM EDTA, 10 mM 2mercaptoethanol, 10% (v/v) glycerol, and dialyzed immediately against 2 liters of the same buffer for 12 h (Fraction IV).

Phosphocellulose
Chromatography-A column of Whatman Pll phosphocellulose (4.9 cmZ x 30 cm) was prepared and washed with 2 liters of 20 mM potassium phosphate buffer (pH 6.51, 0.1 mM EDTA, 10 mM 2-mercaptoethanol, 10% (v/v) glycerol (Buffer B). Fraction IV (50 ml) was adjusted to pH 6.5 by the addition of 50 ml of 20 mM KH2P0,, 10 mM 2mercaptoethanol, 0.1 mM EDTA, 10% (v/v) glycerol and then applied to the column at a rate of 70 ml/h. The column was washed with 360 ml of Buffer B, and the proteins were eluted with a 1.5-liter linear gradient from 0 to 0.5 M KC1 in Buffer B. The gene 4 complementing activity eluted at 0.24 M KCl. Fractions having a specific activity greater than 500 unitslmg were pooled (154 ml), concentrated 4-fold by dialysis against dry polyethylene glycol 6000 for 6 h, and were dialyzed overnight against two 2-liter changes of Buffer A containing 50% (v/v) glycerol. Fraction V (25 ml) was stored at -20" for up to 18 months with no detectable loss of activity. Sephadex G-150 -A column (4.9 cm2 x 100 cm) of Sephadex G-150 was prepared and washed with 1 liter of 20 mM Tris (pH 7.51, 0.1 mM EDTA, 10 mM 2-mercaptoethanol, 10 mM NaCl, 20% (v/v) glycerol at a hydrostatic pressure of 30 cm. Five milliliters of Fraction V were diluted with 5 ml of 20 mM Tris (pH 7.51, 0.1 mM EDTA, 10 mM 2-mercaptoethanol, 10 mM NaCl, layered on the column, and eluted with 20 mM Tris (pH 7.51, 0.1 mM EDTA, 10 mM 2-mercaptoethanol, 10 mM NaCl, 20% (v/v) glycerol. The gene 4 complementing activity eluted at approximately 0.5 column volumes," and the peak fractions, comprising 70% of the activity, were pooled. The pooled fractions were concentrated by applying them to a column (0.64 cm2 x 1.6 cm) of Whatman DE52 DEAE-cellulose that had been equilibrated with Buffer A. The gene 4 complementing activity was eluted with Buffer A containing 0.4 M NaCl, and the fractions containing gene 4 complementing activity were pooled and dialyzed overnight against 1 liter of Buffer A containing 50% (v/v) glycerol (FractionVI).
Fraction VI (1.3 ml) has been stored at -20" for up to 18 months without any appreciable loss of activity (~10%).
Nuclease Activities-The purified gene 4 protein (Fraction VI) contains no detectable exonuclease or endonuclease activities. Incubation of 1 pg of gene 4 protein with 6 nmol of native or denatured T7 L3H]DNA for 20 min at 30" under the conditions of the T7 DNA polymerase stimulation assay (minus T7 DNA polymerase) produced less than 2 pmol of acidsoluble nucleotides. Analysis of the treated DNA by band sedimentation through alkaline sucrose gradients showed no detectable breakdown (<lo%) of the DNA.
DNA Polymerase Activity -Incubation of 1 pg of gene 4 protein (Fraction VI) with native T7 DNA under the conditions of the T7 DNA polymerase stimulation assay resulted in no detectable incorporation (~2 pmol) of [3H]dTl?P into acidinsoluble material. Similarly, incubation of 1 pg of gene 4 protein in the standard T7 DNA polymerase assay (111, in which denatured salmon sperm DNA is the template, resulted in no detectable incorporation (~2 pmol) of L3HldlTP into acid-insoluble material. RNA Polymerase Activity-Although the gene 4 protein, under appropriate cdnditions, can itself catalyze the polymerization of rNTPs to yield short oligoribonucleotides (9),' the gene 4 preparation is not contaminated with either the E. coli or T7 RNA polymerase.
Incubation of 2 pg of gene 4 protein with native T7 DNA, heat-denatured T7 DNA, or 4X viral DNA under conditions of the T7 DNA polymerase stimulation assay (minus T7 DNA polymerase) resulted in no detectable incorporation (~4 pmol) of L3HlUTP or L3'PlATP, UTP, CTP, and GTP into acidinsoluble product.
Nucleoside Triphosphatase Activity-The gene 4 protein is a single-stranded DNA-dependent nucleoside triphosphatase. This activity has been discussed in detail elsewhere (13).

Electrophoretic
Analysis-When Fraction VI of the gene 4 protein was analyzed by electrophoresis on 7.5% polyacrylamide gels under nondenaturing conditions, one major protein band and one minor protein band were observed, with the 3 The presence of purified gene 4 protein decreases the size of a drop causing a decrease in fraction size if fractions are collected by drop counting. A similar effect also occurs in solutions of gene 4 protein containing 0.5 mg/ml of bovine serum albumin. Apparently the gene 4 protein alters the surface tension of solutions. major protein comprising 90 to 95% of the Coomassie-positive material (Fig. la). When an identical unstained native gel was analyzed for gene 4 activity in both the complementation assay and the T7 DNA stimulation assay (see below), both activities had a mobility identical with the protein band that comprised 90 to 95% of the Coomassie-positive material (Fig.  lb).
Again, a major and minor protein band was observed when denatured and reduced samples of Fraction VI of gene 4 protein were analyzed by electrophoresis on 5% polyacrylamide gels in the presence of sodium dodecyl sulfate (Fig.  2a). The major band contained 90 to 95% of the Coomassiepositive material.
The mobility of the gene 4 protein relative to bromphenol blue on 5% polyacrylamide gels containing sodium dodecyl sulfate was 0.64. Comparison of this value with the mobilities of several proteins of known molecular weight (37) yields an apparent molecular weight for the denatured and reduced gene 4 protein of 58,000 + 2,000 (Fig. 2b).
Molecular Weight of Native Gene 4 Protein-The molecular weight of the native gene 4 protein has been determined from the sedimentation coefficient (41) and the Stokes radius by the method of Siegel and Monty (42). The sedimentation coefficient of 3.7 S (Fig. 3b) and the Stokes radius of 30.5 b; (Fig. 3~) indicate a molecular weight of 53,000 ? 6,000. This value is in good agreement with that of 58,000 2 2,000 that was determined for the denatured and reduced gene 4 polypeptide chain and indicates that the gene 4 protein exists as a monomer. 1. Polyacrylamide gel electrophoresis of the gene 4 protein. Samples of gene 4 protein (Fraction VI, 5 pg) were run on parallel gels of 7.5% polyacrylamide under nondenaturing conditions (38). a, One gel was stained with Coomassie blue, and a densitometer tracing was made of the stained gel. b, The second gel was cut into l-mm slices, and each slice was incubated for 2 h at 0" in 0.1 ml of enzyme diluent (see "Experimental Procedures"). Each fraction was then assayed for gene 4 complementing activity (0-O) and T7 DNA polymerase stimulating activity (0-O) as described under "Experimental Procedures." Recovery of both activities from the gel was 8.5%. Gene 4 Protein and T7 DNA Polymerase Together Can Use Duplex DNA Templates -Purified T7 DNA polymerase alone has little or. no activity on duplex DNA templates such as T7 DNA (7,11,12). However, earlier studies with a partially purified gene 4 protein provided strong evidence that the gene 4 protein is essential for DNA synthesis catalyzed by T7 DNA polymerase on duplex DNA templates (71, a property also reported by Scherzinger and Klotz (10). As shown in Fig.  4, the addition of purified gene 4 protein (Fraction VI) to a reaction mixture containing T7 DNA polymerase and native T7 DNA effects a marked stimulation of DNA synthesis. In the experiment shown in Fig. 4, maximum stimulation (greater than 50-fold) occurs at a ratio of 10 gene 4 protein molecules per polymerase molecule. Increasing the amount of T7 DNA polymerase reduces the relative amount of gene 4 protein required for maximum activity (data not shown).
Complementation Activity and T7 DNA Polyrkerase-stimulating Activity Reside in Gene 4 Protein Molecule -As previously reported (71, the complementing activity and the T7 DNA polymerase-stimulating activity purify together during the last stages of the purification described in Table I. During the determination of the sedimentation coefficient of the gene 4 protein described in Fig. 3b, the ratio of the two activities in the protein peak of the sucrose gradient was constant. Finally, as shown in Fig. 1, the two activities migrate together during electrophoresis through polyacrylamide gels under nondenaturing conditions. @he ratio of the complementing activity to T7 DNA polymerase-stimulating activity is identical with the ratio of these two activities in Fraction VI. Synthesis on Duplex DNA is Specific for T7 DNA Polymerase-As previously reported (7) for the less pure gene 4 protein, Fraction VI described here will stimulate only T7  DNA polymerase activity on duplex DNA; gene 4 protein has no effect (<5%) on T4 DNA polymerase or E. coli DNA polymerases I, II, or III.

Specificity
for Duplex DNA-Whereas duplex T7 DNA, isolated from the phage, is a good template for T7 DNA polymerase and gene 4 protein, covalently closed circular duplex PM2 DNA is not (Table II). However, PM2 containing a nick,2 while not an effective template for the T7 DNA polymerase alone, supports synthesis at a rapid rate in the presence of T7 DNA polymerase and gene 4 protein. DNA synthesis on denatured T7 DNA is not dependent on gene 4 protein nor is there any significant stimulation upon the addition of the gene 4 protein (Table II). Although the T7 DNA polymerase alone is more active on linear singlestranded T7 DNA than on duplex DNA, the rate of DNA synthesis, even with gene 4 protein present, is at most only 20% of that catalyzed by the two proteins together on duplex T7 DNA (Table II). The circular single-stranded DNA of 4X174, lacking ends, does not support synthesis by T7 DNA polymerase either in the presence or absence of gene 4 protein.

Requirement for Single Strand Interruptions in Duplex Templates
The results presented in the preceding section show that the T7 DNA polymerase alone can synthesize DNA using linear single-stranded templates, but not circular singlestranded templates or any duplex template. Presumably, the synthesis observed on single-stranded linear DNA reflects the ability of the T7 DNA polymerase to use the 3'-terminus of the polynucleotide strand as a primer and the remainder of the DNA as a template (43). The ability of the T7 DNA polymerase, in conjunction with the gene 4 protein, to use PM2 DNA containing nicks as a template suggests that, like the reaction catalyzed by E. coli DNA polymerase I (43), DNA synthesis by these two proteins requires nicks in the DNA. In this section, we show that DNA synthesis, not only DNA synthesis was measured in the standard T7 DNA polymerase stimulation assay as described under "Experimental Procedures" except that native or heat-denatured T7 DNA (6 nmol), PM2 DNA (4 nmol), and $X174 DNA (4 nmol) were used as templates as indicated, and gene 4 protein was present or absent as indicated. The PM2 DNA with nicks contained one nick per molecule. -In order to determine whether nicks displaying 3'-hydroxyl and 5'-phosphoryl groups in the duplex T7 DNA were contributing to the template activity, we incubated the T7 template DNA with T7 polynucleotide ligase prior to the addition of T7 DNA polymerase and gene 4 protein (Fig. 5). This treatment inhibited the rate of synthesis by 60%.
In order to repair any gaps in the template DNA, the DNA was first incubated with T7 DNA polymerase and the four dNTPs in the presence of T7 polynucleotide ligase (Fig. 5). DNA treated in this manner was only 10% as effective as untreated DNA in promoting synthesis by T7 DNA polymerase and gene 4 protein.
Although not shown here, prior incubation of the T7 DNA with E. coli DNA polymerase I and T4 polynucleotide ligase reduced DNA synthesis catalyzed by the T7 DNA polymerase and gene 4 protein by more than 98%.
In control experiments, heat inactivation of the ligase prior to the addition of T7 DNA polymerase and gene 4 protein did not alter the results shown in Fig. 5 0.3 unit of T7 DNA polymerase and 3.9 units of gene 4 protein per reaction. Molecular weights of 6 x 10" for PM2 DNA (35) and 331 for an average nucleotide were used to calculate the molarity of single strand breaks in a given reaction mixture.
3 min or more after the addition of T7 DNA polymerase and gene 4 protein, no inhibition was observed. Introduction of Nicks into Duplex DNA-Covalently closed circular PM2 DNA will not serve as a template for T7 DNA polymerase and gene 4 protein (<lo/o), while PM2 DNA containing nicks is an efficient template (Table II). PM2 DNA molecules containing one nick per molecule were mixed with covalently closed PM2 molecules in various proportions and then used as template DNA in reactions with gene 4 protein and T7 DNA polymerase (Fig. 6~). The rate of DNA synthesis increased with increasing amounts of template DNA with nicks and is shown in Fig. 6b in the form of a Lineweaver-Burk plot. The K,,, for nicks is 3 x lo-"' M. Assuming that the T7 DNA polymerase is 85 to 90% pure (22) and that all of the DNA polymerase molecules are active, 560 nucleotides are polymerized per min per polymerase molecule in a reaction mixture containing saturating amounts of gene 4 protein.

DISCUSSION
The development of in vitro systems for the replication of bacteriophage T7 DNA has made possible the purification of the gene 4 protein since T7 DNA replication in these systems is dependent on the presence of the gene 4 protein as is replication in uiuo. We, as well as others, have used such complementation assays to purify the gene 4 protein and to study some of its properties (6-9, 13). Although the gene 4 protein can be purified from wild type Escherichia coli infected with wild type T7 phage, we have encountered several problems using this approach.
First, if T7 DNA polymerase or E. coli DNA polymerase I is present in gene 4 fractions, there is some stimulation of DNA synthesis in the complementation assay due to the addition of DNA polymerase alone. To circumvent this problem, an E. coli host carrying a polA mutation and a T7 phage carrying an amber mutation in gene 5 were used to prepare the phage-infected cells in order to eliminate E. coli DNA polymerase I and T7 DNA polymerase activity from the extract.
Second, on several occasions when T7 DNA polymerase was present during purification of the gene 4 protein, we have observed a peak of activity during chromatography that contains both gene 4 protein and T7 DNA polymerase. In fact, the two purified proteins, under appropriate conditions, can be shown to form a complex that can be detected by gel filtration.' Again, the removal of T7 DNA polymerase by a mutation solves this problem.
Third, a major component at all steps of our earlier purification procedure was the T7 DNA ligase. By deleting gene 1.3, the structural gene for the ligase, we have eliminated this protein and its activity. It seems likely that the association of T7 DNA polymerase and DNA ligase with the gene 4 protein during purification is not fortuitous, but further studies on the functional association between these proteins are best carried out with homogeneous proteins. The use of the T7s,,,,,fii,-infected E. coli DllO polA1 endZ and a modified purification procedure to prepare the gene 4 protein has enabled us to obtain a nearly homogeneous protein.
The native gene 4 protein we have purified is a single polypeptide chain with a molecular weight of 58,000. Analysis carried out by Studier (1) of proteins synthesized after T7 infection indicates that two proteins are altered by amber mutations in the gene 4: a major species with a molecular weight of 58,000 and a minor species with a molecular weight of 66,000. Thus, the gene 4 protein we have purified corresponds to the major species produced in a T7 infection. The gene 4 preparation described by Scherzinger et al. (9) contains both the 58,000-and 66,000-dalton components, and they discuss evidence that the two forms have related amino acid sequences. Thus, the 66,000-dalton protein could be a precursor of the 58,000-dalton protein, the latter arising by proteolysis. At present, it is not known whether the 66,000-dalton protein has any of the activities shown to be associated with the 58,000-dalton protein described here. It cannot be ruled out that the larger form of the gene 4 protein performs an as yet unknown role in T7 DNA replication.
The stimulation of T7 DNA polymerase by the gene 4 protein reported here occurs only with duplex DNA templates containing nicks. It should be emphasized here that the requirement for duplex DNA templates occurs only in the absence of rNTPs. In the presence of rNTPs, the gene 4 protein can catalyze the synthesis of oligoribonucleotides on single-stranded linear or circular DNA molecules (9).' Hence, the gene 4 protein can also stimulate synthesis by T7 DNA polymerase on single-stranded DNA by synthesizing oligoribonucleotide primers.