Studies on T3-induced Ribonucleic Acid Polymerase

A procedure for the isolation and purification of bacteriophage T34nduced RNA polymerase from T34nfected Escherichia coli cells is presented. The procedure described leads to the isolation of polymerase preparation that displays a single protein band by polyacrylamide gel electrophoresis and is free of detectable ribonuclease and deoxyribonuclease activities. Other enzyme activities absent from such preparations include inorganic pyrophosphatase, nucleoside diphosphokinase, DNA polymerase, E. coli RNA polymerase, nucleoside triphosphatase, polynucleotide phosphorylase, polyriboadenylate polymerase, and polyphosphate kinase. T3-induced RNA polymerase has been characterized with regard to various biochemical parameters. The enzyme is a single polypeptide chain of molecular weight 105,000 f 5,000. The enzyme activity is completely dependent on the presence of Mg2+, a 11 four nucleoside triphosphates, and T3 DNA. T? DNA shows only 2 to 5 % of activity as that obtained with T3 DNA, while other native DNA preparations examined are inactive. The T3 RNA polymerase is highly sensitive to salt concentrations above 0.03 M or to reagents which react with sulfhydryl groups such as p-hydroxymercuribenzoate and N-ethyhnaleimide. The enzyme catalyzes a T3 DNA-dependent 32PPi exchange into nucleoside triphosphates. However, no pyrophosphorolysis of free RNA can be demonstrated. The “PPi exchange reaction can occur with GTP alone. The exchange reaction with the other three nucleoside triphosphates, present either alone or in combination, does not occur except when GTP is included in the reaction. The effect of E. coli RNA chain termination factor, p, on the T3 RNA polymerase reaction has been studied. It has been found that p factor has no elect on the rate, yield, or size of RNA formed in the T3 RNA polymerase-catalyzed reactions.

The procedure described leads to the isolation of polymerase preparation that displays a single protein band by polyacrylamide gel electrophoresis and is free of detectable ribonuclease and deoxyribonuclease activities.
Other enzyme activities absent from such preparations include inorganic pyrophosphatase, nucleoside diphosphokinase, DNA polymerase, E. coli RNA polymerase, nucleoside triphosphatase, polynucleotide phosphorylase, polyriboadenylate polymerase, and polyphosphate kinase. T3-induced RNA polymerase has been characterized with regard to various biochemical parameters. The enzyme is a single polypeptide chain of molecular weight 105,000 f 5,000. The enzyme activity is completely dependent on the presence of Mg2+, a 11 four nucleoside triphosphates, and T3 DNA. T? DNA shows only 2 to 5 % of activity as that obtained with T3 DNA, while other native DNA preparations examined are inactive.
The T3 RNA polymerase is highly sensitive to salt concentrations above 0.03 M or to reagents which react with sulfhydryl groups such as p-hydroxymercuribenzoate and N-ethyhnaleimide.
The enzyme catalyzes a T3 DNA-dependent 32PPi exchange into nucleoside triphosphates. However, no pyrophosphorolysis of free RNA can be demonstrated. The "PPi exchange reaction can occur with GTP alone.
The exchange reaction with the other three nucleoside triphosphates, present either alone or in combination, does not occur except when GTP is included in the reaction.
The effect of E. coli RNA chain termination factor, p, on the T3 RNA polymerase reaction has been studied.
It has been found that p factor has no elect on the rate, yield, or size of RNA formed in the T3 RNA polymerase-catalyzed reactions. phage T7-infected Escherichia coli cells, reports from this laboratory (1,2) and that of Bautz et al. (4) have described the isolation of a new DNA-dependent RNA polymerase in E. coli infected with bacteriophage T3. The phage-induced polymerase was shown to be physically and biochemically distinct from E. coli RNA polymerase.
The phage polymerase uses only T3 DNA as template and is inactive with native T2, T4, E. co&, calf-thymus DNA, or with poly[d(A-T)].
T7 DNA was shown to be approximately 5% as active as T3 DNA.
In contrast E. coli RNA polymerase is nonspecific in its DNA requirements. T3 RNA polymerase can, however, transcribe poly(dG.dC) to make poly(G) exclusively, which indicates that the polymerase may initiate chains only with GTP.
It has subsequently been shown (2) that RNA chains made by T3 RNA polymerase with native T3 DNA template are initiated exclusively with GTP and that completed RNA chains are released free of template DNA.
Polymerase is also released in this process and, acting catalytically, reinitiates new RNA chains.
The purpose of this communication is to report the purification and characterization of the T3-induced RNA polymerase and various properties of the polymerase reaction.
Such enzyme preparations have already been used to study various aspects of the transcription process catalyzed by T3 RNA polymerase (2,5). The accompanying paper (6) describes the reaction of the purified enzyme with denatured DNA templates.

Materials
Preparation of TS-Bacteriophage Lysate and TS Phage-infected cells-The growth medium (Medium A) contained, per liter, IO g of glucose, 10 g of Casamino acids, 0.1 g of yeast extract, 0.3 g of MgSO+ 5 g of NaCl, 2 g of NH&l, 0.03 g of CaC12, 7 g of K2HP04, and 3 g of KH2P04.
For the preparation of crude T3 (wild type) lysates, E. coli B (strain Sy 106) was grown aerobically at 37". When the optical absorbance of growing bacteria reached 0.5 to 0.6 at 660 nm, the culture was infected with phage T3 at a multiplicity of 0.2. After lysis occurred (30 to 40 min), 1 ml of CHC13 and 24 g per liter of NaCl were added, and the lysate was centrifuged for 10 min at 5000 X g to remove debris (crude bacteriophage lysate).
For the preparation of T3 phage-infected cells, E. coli Sy 106 were grown at 30" aerobically in Medium A. When the optical absorbance of bacteria reached 0.8 at 660 nm, the culture was infected with phage T3 at a multiplicity of 8. Sixteen minutes after infection, the culture was rapidly cooled at 5", and the cells were harvested in a Sharples continuous flow centrifuge which was cooled to 5". The infected cells were stored at -40" until use. Under the above condition of infection, cell lysis was complete after 28 min.
Isolation of DNA-T3 phage was further purified from the crude bacteriophage lysate by precipitation with polyethylene glycol followed by extraction of the precipitated material with 1 M KC1 solution as described by Yamamoto et al. (7). Subsequently, bacteriophage was purified by isopycnic banding in CSCI.
Incubation was for 30 min at 37", after which reactions were terminated by adding 0.5 ml of 0.2 M EDTA, 0.2 ml of 0.1 M PPi, pH 6.0, and 0.2 ml of a 10% suspension of Norit.
The reaction mixtures were gently mixed in ice for 5 min, followed by addition of 2 ml of 0.01 M PPi (adjusted to pH 6.0 with KH~POI), and were centrifuged for 5 min at 10,000 x g. The Norit pellet was suspended in 2 ml of 0.01 M PPi, pH 6.0, filtered through a glass fiber filter (Whatman GF/C soaked with 0.01 M PPi, pH 6.0) and washed four times with 3-ml aliquots of 0.01 M PPi, pH 6 (10,11). DEAE-cellulose (DE-52) and phosphocellulose (P-11) were from Whatman. Bovine serum albumin (twice recrystallized, obtained from Pentex Inc.) was dialyzed against 1 mM EDTA for 48 hours followed by dialysis against water for 24 hours according to Singer (12). E. coli RNA polymerase was obtained as described previously (13). Protein factor p was isolated and purified from E. coli MRE-600 as described by Roberts (14). Incubation of 5 pg of p factor with either 10 nmoles of f2 bacteriophage [3H]RNA or 10 nmoles of T7 t3H]DNA for 30 min at 37" did not alter the sedimentation profile of either polymer, indicating the absence of detectable DNA or RNA endonucleases in p factor preparations.

Methods
Assay of TS RNA Polymerase-The conditions of the assay, unless otherwise specified, were as follows.
The polymerase reaction was initiated by the addition of T3 RNA polymerase (2 to 10 units).
During enzyme purification, 1 pg of rifampicin was also included in the reaction mixtures in order to inhibit E. cloi RNA polymerase activity.
Incubation was for 15 min at 37" and reactions were terminated by addition of 3 ml of ice-cold 5% CCl&OOH; after 5 min at O", mixtures were filtered through nitrocellulose membrane filters (Millipore filters).
The filters were washed four times with 3-ml aliquots of 5% CC13COOH, dried under gentle heat, and their radioactivity content measured in a liquid scintillation counter. The column was then washed with the above buffer solution, and 0.5.ml fractions were collected.
The labeled RNA product emerged from the column as a single symmetrical peak. Under these conditions over 99% of the 32P present in the RNA product was precipitated with 5% CCl&OOH.
With native T3 DNA, T3 RNA polymerase produces RNA chains free of template DNA (2), while with single-stranded DNA templates (e.g. 4X-174 DNA), the RNA produced is present as DNA-RNA hybrids (6). Other Melhods-Protein was determined by the turbidometric method of Biicher (16).

Purijication of TS RNA Polymerase
Unless otherwise indicated, all operations were carried out at 04" and all buffer solutions contained 5 x 1O-4 M EDTA and 10d2 M 2-mercaptoethanol.
The purification procedure is described below, while the results of a typical preparation from 30 g of T3-infected E. coli cells are summarized in Table I. In this assay, 1 unit of T3 RNA polymerase activity is equal to the incorporation of 1 nmole of ['4C]UMP into RNA in 15 min at 37".
Assay for PPi Exchange-The PPi exchange reaction was measured by determining the amount of 32PPi incorporated into a Norit adsorbable form as described by Krakow and Fronk (15). Unless otherwise specified, reaction mixtures (0.2 ml)

Preparalion
of Crude Extracts-Thirty grams of partially thawed T3-infected E. coli cells were ground with 60 g of Alcoa alumina powder A-301 in a precooled mortar until a fine paste was obtained.
The mixture was then extracted with 90 ml of extraction buffer which contained 25 mM Tris-HCl buffer, pH 7.8, 10 mM MgC&, and 30 mM NH&l (Buffer A). The suspension was centrifuged for 15 min at 23,000 X g in a Sorvall centrifuge. The supernatnat fluid was decanted, and the pellet was extracted with 30 ml of Buffer A. The combined supernatants from the two extractions were recentrifuged in the Sorvall as de-  To each pellet were added 5 ml of Buffer A containing 1 M NH&l. The pellets were gently homogenized in a glass homogenizer and the suspension was incubated at 4" for 1 hour. The suspension was then centrifuged for 15 min at 30,000 x g, and the supernatant fluid was recentrifuged for 3 hours at 150,000 x g. The supernatant fluid was adjusted to 30% saturation by the addition of solid ammonium sulfate (165 g per liter).
After 20 min of gentle stirring, the precipitate was removed by centrifugation at 30,000 x g for 15 min, and the supernatant fluid was adjusted to 50% saturation with solid ammonium sulfate (130 g per liter). After 30 min of gentle stirring, the precipitate containing T3 RNA polymerase was collected by centrifugation at 80,000 x g for 20 min in a Spinco rotor No. 30, and dissolved in 5 ml of 50 mM Tris-HCl, pH 7.8, containing 10% glycerol. The solution was dialyzed for 6 hours against 1 liter of 50 mM Tris-HCl, pH 7.8, containing 10% glycerol (ammonium sulfate I fraction).
Ammonium Sulfate II-Ammonium sulfate I fraction (10 ml) was diluted with an equal volume of buffer containing 10 mM Tris-HCl, pH 7.8, and 0.5 M KCl. Fifty grams of DEAEcellulose, wet weight, previously equilibrated with 20 mM Tris-HCl buffer, pH 7.8, containing 0.25 M KCl, were added.
The suspension was stirred for 20 min and was subsequently applied to a column of DEAE-cellulose (1.4 X 30 cm) equilibrated with 20 mM Tris-HCl, pH 7.8, and 0.25 M KCl. The column was then washed with the same buffer solution containing 5y0 glycerol. Under these conditions most proteins were unretarded, while nucleic acids were retained.
The elution of protein from the column was followed by measuring the absorbance of the effluent at 280 nm. Fractions containing the bulk of the unretarded protein were pooled and treated with saturated ammonium sulfate solution, pH 7.5, to 55% saturation.
After stirring for 30 min, the precipitated protein was collected by centrifugation (80,000 x g for 15 min in a Spinco No. 30) and dissolved in a 4 ml of a buffer containing 50 RIM Tris-HCl buffer, pH 7.8, and 5% glycerol (ammonium sulfate II fraction).
Sephadex G-800 Gel Filtration-The ammonium sulfate II fraction obtained as described above was layered onto a column of Sephadex G-200 (2 X 100 cm) previously equilibrated with 50 mM Tris-HCl buffer (pH 7.8) containing 5y0 glycerol.
The column was developed with the same buffer, and fractions of 5 ml were collected.
Protein was monitored by measuring A280 of the effluent. Fractions containing the bulk of the T3 RNA polymerase activity which showed at least a a-fold increase in specific activity relative to the ammonium sulfate II fraction were pooled (Sephadex G-200 eluate, Fraction III).
DEAE-cellulose Chromatography-The Sephadex G-200 eluate was added to a DEAE-cellulose column (1 x 20 cm), previously 6639 equilibrated and washed with 50 mM Tris-HCl, pH 7.8. The column was washed with 30 ml of 50 mM Tris-HCl, pH 7.8, followed by 60 ml of the same buffer but also containing 0.04 M KC1 and 5% glycerol.
A linear gradient of 100 ml total volume from 0.05 M Tris-HCl (pH 7.8) + 0.04 M KC1 to 0.05 M Tris-HCl (pH 7.8) + 0.15 M KC1 was then applied to the column.
The buffer solutions also contained 5% glycerol.
Fractions of 3 ml were collected.
T3 RNA polymerase activity was eluted at a KC1 concentration between 0.07 and 0.1 M. Fractions containing T3 RNA polymerase activity were pooled (DEAEcellulose eluate, Fraction IV).
Phosphocellulose Chromatography-Fraction IV was immediately applied to a phosphocellulose column (2 x 8 cm), previously equilibrated with 50 mM Tris-HCl, pH 7.8. The column was then washed with 10 ml of the above buffer, followed by 30 ml of 0.1 M potassium phosphate buffer, pH 7.5, containing 55/ glycerol.
A linear gradient of 60 ml total volume from 0.1 M to 0.4 M potassium phosphate buffer (pH 7.5) containing 10% glycerol was applied.
The active fractions were pooled and were dialyzed for 6 hours against 500 ml of 0.1 M Tris-HCl buffer (pH 7.8) and 50% glycerol.
This dialysis procedure resulted in a concentration of the pooled fraction. The dialyzed fraction (phosphocellulose eluate, Fraction V) was stored at -20".

Properties of PuriJied Enzyme
Stability-T3 RNA polymerase was routinely stored at the Fraction V stage at -20".
Under these conditions, less than 15% loss of polymerase activity occurred over a 2-month period. We have also added dialyzed bovine serum albumin to Fraction V to a concentration of 1 mg per ml of albumin. Under these conditions the enzyme preparation was found to be stable for over 6 months at -20" or at -90". Repeated freezing and thawing or storage at 0" caused marked loss of enzyme activity; over 60 to 70% of enzyme activity was lost in less than 2 weeks.
Purity-When 20 Hg of the Fraction V enzyme were analyzed by native polyacrylamide gel electrophoresis according to the method of Maize1 (17) at pH 8.0, a single band migrating toward the anode was detected.
However, we have not tried to show that the enzymatic activity coincides with this protein band. (No protein bands were observed migrating towards the cathode at pH 4.5.) Purified polymerase (Fraction V) was also analyzed by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate (Fig. 1). The single band observed provided further evidence of homogeneity and also indicates that T3 RNA polymerase is composed of a single type of polypeptide chain.

Endow&ease
Activity towards RNA and DNA-Incubation of 10 nmoles of either f2 bacteriophage [3H]RNA or T7 t3H]DNA with 40 units of T3 RNA polymerase for 1 hour did not alter the sedimentation profile of either polymer, indicating the absence of RNA and DNA endonucleases in the T3 RNA polymerase preparation (Fig. 2) Exonuclease Activity-When azP-labeled f2 bacteriophage RNA (20 nmoles, specific radioactivity, 1 x lo5 cpm per nmole) or T3 RNA polymerase product (20 nmoles, specific activity, 2.5 x lo5 cpm per nmole) or 3H-labeled native or heat-denatured T7 DNA was incubated with 100 units of T3 RNA polymerase for 16 hours at 37", less than 0.1 y0 of radioactivity was converted to the 5% CClzCOOH-soluble form. The experimental conditions used were similar to those described in a typical T3 RNA T3 RNA polymerase (Fraction V, 5 pg) was subjected to polyacrylamide gel electrophoresis (6.5y0) in the presence of sodium dodecyl sulfate at pH 7.0 according to the methods described by Maize1 (17). The gels were stained with Coomassie blue and destained electrophoretically.
The authors are indebted to Dr. P. Gupta of this institution for considerable help in the gel electrophoresis experiments.
FIG. 2 (right). Sedimentation profile of f2 bacteriophage RNA and T7 DNA after treatment with T3 RNA polymerase.
In Experiment A, 10 nmoles of f2 [SH]RNA (15,000 cpm for nmole) and in Experiment B, 14 nmoles of T7[3H]DNA (31,000 cpm per nmole) were added. The T7 DNA was heated at 70" for 4 min and fa& cooled before use to insure disaggregation.
Reaction mixtures were incubated at 30" with 40 units of T3 RNA uolvmerase for 30 min. In the experiment presented in A, the &a&on was terminated by the addition of EDTA to 0.04 M and sodium dodecyl sulfate to 0.5%. The reaction mixture was stored at 0" for 15 min and the precipitate formed was removed by centrifugation.
The supernatant containing RNA was subjected to formaldehyde treatment to disrupt secondary structure by the following procedure.
Sodium phosphate buffer, pH 7.7 (final concentration 0.1 M), and formaldehyde to 3$$ (v/v) were added and the mixture was heated at 63" for 15 min followed by cooling to 20" according to Boedtker (18). Aliquots were layered onto 5 ml of 5 to 20% sucrose gradients.
Gradients contained 0.1 M sodium phosphate buffer, pH 7.7,1.1 M formaldehyde, and 0.2y0 sodium dodecyl sulfate. Centrifugation was for 4 hours at 48,000 rpm in the SW 50.1 rotor at 20". The acid-insoluble radioactivity was measured in fractions (0.15 ml) collected from the bottom of the tube. In the experiment presented in B, the reaction was terminated by the addition of 0.06 ml of a solution containing 1.15 M NaOH, 0.3 M EDTA, and 1.85 M NaCl. The entire mixture was layered onto a 5 to 20y0 alkaline sucrose gradient (containing 0.7 M NaCl, 0.3 M NaOH, and 10-a M EDTA) and centrifuged in an SW 50.1 rotor for 165 min at 49,000 rpm at 5". Fractions were collected directly into Bray's solution from a hole pierced in the bottom of the tube and counted in a Packard Tri-polymerase reaction mixture except that all four nucleoside triphosphates were omitted, and 32P-labeled RNA or 3H-labeled T7 DNA replaced T3 DNA.
Assay for Other Enzyme Activities in Purijied TS RNA Polymerase-Measurement of the following enzyme activities was carried out under conditions described for the routine T3 RNA polymerase assay (see "Methods") except that DNA and ribonucleoside triphosphates were omitted and incubation was for 2 hours at 37" with 40 units of T3 RNA polymerase (Fraction V). These enzyme activities were assayed as described previously (13).
Nucleoside Triphosphatase-T3 RNA polymerase incubated with 25 nmoles of any one of the four y-32P-labeled ribonucleoside triphosphates (specific activity, 2 x log cpm per pmole) did not yield detectable Norit-nonadsorbable 32P ( < 1 pmole). DNA Polymerase-The assay utilized dAT copolymer as primer and was carried out as described by Richardson et al. (19) with [aH]TTP (2 x lo* cpm per pmole) as the labeled deoxynucleoside triphosphate.
Molecular Weight-The sedimentation constant of T3 RNA polymerase was found to be 5.4 by glycerol gradient centrifugation; in these experiments, marker proteins were run in parallel gradients (Fig. 3). Assuming that T3 RNA polymerase is a typical globular protein (5 = 0.73, f/f0 = 1.25), the molecular weight based on the sedimentation value is 105,000 f 5,000. The subunit molecular weight of T3 RNA polymerase was determined by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate.
The mobilities of T3 RNA polymerase relative to several marker proteins were measured.
As shown in Fig. 4, the molecular weight obtained in this way was 105,000 =t 5,000. Since the molecular weight of the polymerase was the same under native as well as denaturing conditions, T3 RNA polymerase probably consists of a single polypeptide chain. of the other three ribonucleoside triphosphates (Fig. 7). The apparent K, for UTP was calculated to be 0.75 x 10F4 M. Similar results were obtained for ATP (apparent K, = 1 X 1O-4 M) and for CTP (apparent K, = 0.8 X 10M4 M) (data not shown).
The effect of GTP concentration on polymerase activity is shown in Fig. 8. It is evident that there is a lag in nucleotide incorporation at low concentration of GTP. By extrapolation of the linear portion of the double reciprocal plot, as indicated in Fig. 8B, to the abscissa, the apparent K, for GTP was calculated to be 1.  of the data presented in Fig. SB, obtained at low concentration of GTP, was plotted as l/V versus 1/S2, a linear curve was obtained (Fig. SC). It is known from previous studies (25) that the first two nucleotides at the 5' end of RNA chains synthesized with T3 RNA polymerase are both guanine. Thus, the sequence at the 5' end of RNA chains is pppGpGp. . . . If the ratelimiting step in RNA synthesis is the formation of the first phosphodiester bond (initiation reaction) through a bimolecular reaction between 2 GTP molecules, the rate of RNA synthesis will be second order with respect to GTP concentration. Thus a plot of l/V versus l/S2 will give a linear curve at low concentrations of GTP.
With other nucleoside triphosphates which are only involved in sequential elongation of nucleotides to the 3' end of the nascent RNA chain, the rate of RNA synthesis will be expected to be a first order reaction with respect to substrate concentration.
Thus for these three triphosphates l/V versus l/S should give a straight line. This is what was obtained for UTP, CTP, and ATP (data for UTP are shown in Fig. 7B).
T2 DNA or calf thymus DNA.
With T7 DNA, however, a slower rate of exchange reaction relative to T3 DNA occurred. Thus the specificity for template DNA for the pyrophosphate exchange reaction is similar to that for the polymerization reaction catalyzed by T3 RNA polymerase.
The ribonucleoside triphosphate requirements for pyrophosphate exchange reactions are shown in Table IV.
It is evident that GTP is absolutely required for the 32PPi exchange reaction.
This result is in keeping with the observat,ions reported previously from this laboratory that in the T3 RNA polymerase reaction, RNA chains are initiated exclusively with GTP (2). It should be noted (Table IV)     Two reaction mixtures were prepared; one contained 20 nmoles of GTP as the only ribonucleoside triphosphate, and the other contained 20 nmoles of each of the four ribonucleoside triphosphates. The other conditions of the two reaction mixtures were as described under "Methods." The specific activity of 32PPi was lo5 cpm per nmole and 12 units of enzyme were used. After incubation at 37" for 30 min, the reaction was terminated and the nucleotides were adsorbed to Norit as described under "Methods." The Norit was stashed three times by suspending it in 2 ml of 0.01 M PPi, pH 6.0, and the supernatant fluid was removed each time by IOU-speed centrifugation.
The nucleotides were then eluted from Norit with 1 N NH, in 507, ethanol solution. The eluate was evaporated to dryness and suspended in a small volume of HTO. After adding ATP, UTP, CTP, and GTP (2 pmoles of each) to both reaction mixtures as carrier, each solution was applied to a column of Dowex 1 (Cl) (1 X 10 cm, 100 to 200 mesh, 2% cross-linked) which had been extensively washed with water. Each column was sequentially developed with 100 ml of each of the following reagents in the following order: (a) 10 mM HCI; (5)  The elution profiles of the nucleotides were determined by measuring the absorbance of the effluents at 260 and 280 nm. The above separation procedure completely resolved CTP, ATP, GTP, and UTP which were eluted sequentially in this order from the column. The fractions Were analyzed for 3ZP. The total 32P radioactivity in each of the nucleoside triphosphate region was calculated. The recovery of 32P from each chromatogram was greater than 90%.  (26)(27)(28).
At low ionic strength (<O.l M KCl), however, the presence of a protein factor, p, is necessary to bring about termination and release of RNA chains (14). Polymerase is not released and reinitiation of RNA chains does not occur (14,28,29). The over-all effect of p factor on transcription of DNA catalyzed by E. coli RNA polymerase can be summarized as follows.
(a) Addition of p factor to an E. coli RNA polymerase reaction depresses RNA synthesis without affecting the initial rate of RNA synthesis.
(5) RNA chains formed in the presence of p factor are smaller and more homogeneous in size than are those formed in its absence. It was of interest to investigate whether p factor affected in &TO RNA synthesis catalyzed by T3 RNA polymerase. Incubation was for 20 min at 37" followed by the addition of 5% CCl&OOH to terminate the polymerase reaction.
The As shown in Table VI, p factor had no effect on either the rate or extent of RNA synthesis catalyzed by T3 RNA polymerase. The concentration of p factor used in this experiment was sufficient to cause over 50% depression of the rate of nucleotide incorporation catalyzed by E. coli RNA polymerase. Variation of the concentration of p factor in a T3 RNA polymerase reaction over a 20-fold range (1 to 20 pg) also had no effect on RNA synthesis catalyzed by T3 RNA polymerase (data not shown). directed) RNA synthesis by E. coli RNA polymerase (Table   VII).
In these experiments, p factor was first incubated with T3 RNA polymerase in the absence of nucleoside triphosphates, and the reaction was initiated by addition of E. coli RNA polymerase and nucleoside triphosphates.
Since T3 RNA polymerase is specific for T3 DNA and does not copy T4 DNA, nucleotide incorporation under these conditions is solely a measure of the E. coli RNA polymerase activity. As shown in Table   VII, the extent of inhibition of the E. coli RNA polymerase activity by p factor was identical whether or not p factor was preincubated with T3 RNA polymerase. In addition, the presence of a large excess of T3 RNA polym- The effect of p factor on the size of RNA produced by either erase had no effect on the p-mediated inhibition of (T4 DNA-E. coli or T3 RNA polymerase was studied (Fig. 9). In agree- Polymerase reaction mixtures (0.5 ml) were prepared as described under 'LMethods" except that 10 mM MgCl, was used in the E. coli RNA polymerase system and [W]UTP (1 X 107 cpm per pmole) was added. Other additions were 6 units of T3 RNA polymerase or E. coli RNA polymerase and 1 rg of p factor where indicated.
Incubation was at 37" and 0.05-ml aliquots were removed at indicated times and treated with 3 ml of ice-cold 5yo CCl&OOH.
The amount of '%-radioactivity incorporated into an acid-insoluble RNA product was determined by filtration on Millipore membranes as described under "Methods." The amount of p factor added in these experiments was a saturating amount for the E. coli RNA polymerase system.  QA). In contrast, p factor had no effect on the average size of RNA formed from T3 DNA template transcribed by T3 RNA polymerase (Fig. QB) . Similar results were obtained with T7-induced RNA polymerase by Goldberg and Hurwitz (30). DISCUSSION The present communication extends previous studies on the T3-induced RNA polymerase (1,2). In this paper, we have reported studies on the detailed purification procedure and physical and chemical characterization of T3 RNA polymerase as well as studies on requirements of the polymerase reaction.
The final enzyme preparation as examined by sodium dodecyl suIfate-polyacrylamide gel electrophoresis displayed a single protein band. The enzyme preparation has been examined for nuclease contamination and, within the limitations of the assays employed, appears to be free of RNase and DNase activities. Many other known enzyme activities involving nucleoside triphosphates were also absent. This template specificity distinguishes T3 RNA polymerase from T7 RNA polymerase described by Chamberlin et al. (3). In contrast, in the DNAdependent 32PP; exchange reaction with nucleoside triphosphates catalyzed by T3 RNA polymerase, T7 DNA was nearly 30% as active as T3 DNA (Table III).
A comparison of the biochemical properties of T7 and T3 RNA polymerases (l-4) shows that the two polymerases behave similarly.
Both T3 and T7 RNA polymerases are single polypeptide chains of molecular weight approximately 105,000 f 5,000. Both enzymes have similar requirements for RNA synthesis.
Both enzymes are insensitive to rifampicin but sensitive to salt concentrations above 0.03 M. Both enzymes, in addition to transcribing the homologous DNA, work efficiently with the homopolymer pair poly(dG.dC) to make poly(G) exclusively.
This result suggests that promoter sites recognized by these two polymerases may be dC-rich regions of the DNA. However, the T3 RNA polymerase appears to have a more restricted initiation site, since this enzyme will only use native T3 DNA and is virtually inactive with T7 DNA.
The T7 RNA polymerase, however, can utilize native DNA from either T3 or T7 as template (4).
T3 RNA polymerase catalyzes a T3 DNA-dependent PPr exchange reaction with nucleoside triphosphates.
The requirements for the PPi exchange reaction were found to be the same as that observed for RNA synthesis except that the 32PPi exchange reaction can occur with GTP alone. In the absence of  9. Effect of p factor on size of RNA produced by Escherichia coli and T3 RNA polymerases.
Two sets of polymerase reaction mixtures with T3 DNA were prepared, one with 3 units (3 nmoles) of E. coli RNA polvmerase (A) and one with 6 units of T3 R,NA polymerase (B). 'Reaction mixtures (0.5 ml) in both sets contained the same comnonents as described in leaend to Table VI except that 1 pg of p facior was added where indicated.
[3H]UTP (3ooO cpm per nmole) was the labeled nucleotide used. After incubat,ion at 37" for 20 min, the polymerase GTP, the exchange of 32PPi with any one of the other three ribonucleoside triphosphates ATP, UTP, and CTP, present either alone or in combination, does not occur. This result is in keeping with observations reported previously from this laboratory that RNA chain initiation by T3 RNA polymerase proceeds exclusively by the incorporation of an intact GTP to form the 5' terminus of the RNA (2). The second nucleotide incorporated by forming a phosphodiester bond with the initiating GTP has also been identified to be guanine.
Thus the sequence at the 5'-triphosphate end of RNA chains is pppGpGp . . . (25). The observation that the PPi exchange reaction can occur with GTP alone is most readily explained as a consequence of a polymerization step to form pppGpG followed by pyrophosphorolysis of the newly formed phosphodiester bond.
Results presented in this paper also demonstrate that transcription of T3 DNA by T3 RN.4 polymerase is unaffected by the E. coli RNA chain termination factor, p. It has been shown with E. coli RNA polymerase that the presence of p factor is necessary to bring about termination and release of RNA chains (14).
Addition of p factor to an E. coli RNA polymerase reaction mixture causes inhibition of RNA synthesis without affecting the initial rate of RNA synthesis.
p appears to interact with E. coli RNA polymerase at specific DNA sites leading to the formation of RNA chains of well defined length (30). The addition of p factor to T3 RNA polymerase reaction, on the other hand, is without effect on either the rate, yield, or size of RNA produced. Similar results have been obtained by Dunn et al. (35).
The inability of p factor to affect transcription by T3 RNA polymerase may be due to the absence of signals for p-mediated termination in the regions of the T3 genome which are transcribed by T3 RNA polymerase.
Alternatively, and perhaps more likely, p may be specific for E. coli RNA polymerase and be unable to interact with T3 polymerase under any condition. Termination of in vitro RNA synthesis has been shown to occur in the T3 RNA polymerase reaction with the release and reinitiation of RXA chains from the T3 DNA template (2). If RNA produced in vitro by T3 RNA polymerase is found to terminate improperly when compared to in viuo RNA formed in T3-infected cells, we must consider the possibility that there exists a phage-coded termination factor specific for T3 RNA polymerase.
Finally, we have calculated that with native T3 DNA template and under the condition of the assay employed for T3 RNA polymerase in this paper, the rate of RNA synthesis per mole of T3 RN-4 polymerase is approximately 10 times faster than that obtained per mole of E. coli RNA polymerase. ' The above calculation indicates a higher degree of efficiency of catalytic function for T3 RNA polymerase than for E. coli RNA polymerase.
The biological role of T3 RNA polymerase is clear. RNA products made in vitro by purified T3 RNA polymerase have been shown to contain all of the sequences present in "late" mRNA isolated from T3-infected cells (5,35). Thus the enzyme is responsible for transcription of late mRNA during the life cycle of bacteriophage T3.