The Purification and Characterization of Adenosine Triphosphate-Ribonucleic Acid Adenylyltransferase from Pseudomonas putida*

SUMMARY ATP-RNA adenylyltransferase was isolated from the 80,000 X g pellet of Pseudomonas putida. The enzyme specifically catalyzed the incorporation of AMP from ATP into a polymeric product. The of incorporation and not the The reaction upon the cofactor, neither be replaced nor supplemented manganese ion. The polymerization reaction, which the pyrophosphate,

Numerous enzymes have been reported in the past few years which catalyze the addition of the nucleoside monophosphate portion of a nucleoside triphosphate onto the 3'-end of oligoor polynucleotides.
The terminal addition of deoxyribonucleotides to oligo-or polydeoxyribonucleotides has been well documented. Krakow, Coutsogeorgopoulos, and Canellakis (1) reported the isolation of a terminal deoxyribonucleotide transferase from the nuclear fraction of calf thymus gland which utilized each of the four common deoxyribonucleoside triphosphates and, in the presence of Mg2+, added the corresponding monodeoxyribonucleotides onto the 3'-end of denatured DNA. The enzyme, which was distinct from DNA polymerase, synthesized short chains of deoxyribonucleotide product (less than 4 residues in length).
Referred to as the "nuclear terminal addition enzyme" (2), it was also capable of incorporating the common ribonucleoside triphosphates, but only a single ribonucleotide residue could be added to each primer molecule (2, 3).
The isolation of another terminal deoxyribonucleotide transferase has been reported by Bollum (4), who utilized the soluble fraction of calf thymus gland.
This enzyme, distinct from the nuclear enzyme and referred to as "polydeoxynucleotide synthetase" (5), incorporated the mononucleotide residues of each of the four common deoxyribonucleoside triphosphates onto the 3'-end of a denatured DNA or oligodeoxyribonucleotide primer. Variable chain lengths (one to 600 nucleotides), random copolymerization of deoxyribonucleotides, and synthesis de nouo of polydeoxyribonucleotides could be achieved depending upon incubation conditions (5-7).
Enzymes which catalyze the terminal incorporation of ribonucleotides from ribonucleoside triphosphates into RNA have been shown in a wide variety of systems. One of these enzymes is the well characterized and ubiquitous ATP(CTP)-tRNA nucleotidyltransferase, which synthesizes the pCpCpA sequence on the 3'.end of transfer .
This enzyme is specific for transfer RNA and produces a product of well defined length and sequence, factors which distinguish it from other terminal ribonucleotide-incorporating activities. The remainder of these terminal ribonucleotide-incorporating activities, which have been isolated from mammalian (13-25), plant (26), avian (27)(28)(29), and bacterial (30)(31)(32)(33) sources, have not been as well characterized.
Nevertheless, all require a ribonucleoside triphosphate as substrate, a divalent metal ion as cofactor, and an oligo-or polyribonucleotide as primer. In those cases in which sufficient data are available, the reactions appear to proceed, with one possible exception (see below), by the addition of monoribonucleotide residues onto the 3'-end of the primer.
The resulting products are homoribopolymer chains of varying lengths (one to 200 nucleotides), covalently attached to the primer molecule.
Within these basic characteristics, however, a great deal of diversity exists among the various ribonucleotide incorporating activities.
Enzymes which are reasonably specific for each of the four common ribonucleoside triphosphates have been prepared from Escherichia coli (ATP), calf thymus (CTP), rat liver (UTP), and spinach (GTP), while an enzyme fraction Issue of nlarch 25,1970 K. J. Payne and J. A. Boezi 1379 from Landschutz ascites tumor cells will utilize any of the ribonucleoside triphosphates. It is not known whether the latter enzyme fraction contains one enzyme which will incorporate each ribonucleoside triphosphate or an enzyme for each one.
Among the best characterized of the ribonucleotide-incorporating activities are those from E. co& In 1962, two groups of investigators reported the isolation of specific ATP-incorporating activities from this source. These enzymes were isolated by Gottesman,Canellakis,and Canellakis (30) using the soluble fraction and August,Ortiz,and Hurwitz (31) using the ribosomal pellet. In the presence of the divalent metal ions, Mg2+ and Mn2f, the enzyme from the soluble fraction was reported to catalyze the terminal addition of adenylate residues onto the 3'-end of a RNA primer, producing chains of 25 to 35 nucleotides in length.
The enzyme isolated from the ribosomal pellet, referred to as "polyriboadenylate polymerase" (34), also required ATP and an RNA primer.
However, the reaction, in the presence of Mg2+ alone, reportedly produced chains synthesized de novo of poly A, 100 to 200 nucleotides in length.
Recently, Hardy and Kurland (32) also described an ATP-incorporating activity from E. coli ribosomes.
This enzyme, which required both Mg2+ and Mn2+ for optimal activity, did not produce synthesis de novo of poly A but instead added adenylate residues onto the 3'.end of a ribosomal RNA primer.
The reason for the discrepancy between the results of August, Ortiz, and Hurwitz and Hardy and Kurland is not known.
Perhaps there are two different ATP-incorporating enzymes present in E. coli ribosomes.
Possibly there is a single enzyme which, depending on the conditions, is able to catalyze the synthesis de novo of poly A or the addition of adenylate residues to the 3'.end of an RNA primer.
It should be pointed out that the purification procedures used by the two groups were different, the specific activities of the partially purified enzymes varied, and neither enzyme was free of endogenous RNA.
The purpose of this report is to present the purification and characterization of an enzyme, designated ATP-RNA adenylyltransferase, from the ribosomal pellet of Pseudomonas putida. The enzyme catalyzes the incorporation of adenylate residues from ATP into a polymeric product, with the concomitant release of inorganic pyrophosphate.
The reaction is completely dependent upon exogenous RNA and requires the cofactor, magnesium ion. The product is a homopolymer chain of adenylate residues, greater than 100 nucleotides in length, which is covalently attached to the 3'-end of the added ribosomal RNA primer.
Calf thymus DNA, poly A, poly U, poly C, poly I, and phosphoglucomutase were purchased from Sigma. Labeled nucleotides were obtained from Schwarz BioResearch.
P. putidu bacteriophage gh-1 DNA (35,36) was purified by the method of Thomas and Abelson (37). The gh-1 and calf thymus DNA were denatured by heating at 100" for 10 min followed by quick cooling.
UDPG pyrophosphorylase was isolated from calf liver (38)  The growth media contained, per liter, 5 g of bacto yeast extract (Difco), 6 g of (NH,),HPO+ 3 g of KH2PO4, 8 g of NaCl, 2.1 g of MgClz.6H20, 12 g of glucose, and 5 mg of FeC13. Bacterial growth was continued until the early stationary phase, at which time the culture was harvested by means of a Sharples centrifuge.
Cells were stored at -20" for several months without detectable loss of activity.
Preparation of Ribosomal and Soluble RNA-Frozen P. putidu cells were disrupted as described under "Results ("Purification of Enzyme")" and the initial 80,000 X g pellet was resuspended in buffer containing 10 mM Tris-HCl (pH 8.0), 10 mM MgC12, and 10 pg per ml of DNase (Worthington, electrophoretically purified). After incubation at 37" for 10 min, the ribosomes were collected by centrifugation at 80,000 x g for 120 min. The pellets were then resuspended in 10 mM magnesium acetate-0.2% SDS1 (pH 5.1) by stirring at 3" for 90 min. An equal volume of water-saturated phenol was added, and, after stirring for 15 min, the suspension was centrifuged at 27,000 x g for 10 min. The aqueous layer was removed and saved and the remaining material was re-extracted with the magnesium acetate-SDS buffer. The new aqueous layer was combined with the first and the combined aqueous layers were treated with an equal volume of water-saturated phenol. The aqueous layer was separated and again treated with phenol. The final aqueous layer was made 0.3 M with respect to sodium acetate and ribosomal RNA was precipitated by the addition of 2 volumes of cold ethanol. The precipitate was dissolved in 10 mM Tris-HCl (pH 8.0). The solution was then made 0.3 M with respect to sodium acetate and ribosomal RN,4 was reprecipitated with cold ethanol. This final precipitate was redissolved, dialyzed extensively against 10 mM succinate-NaOH (pH 5.3), and stored at -20".
The resulting ribosomal RNA was free of detectable small molecular weight RNA as determined by sucrose density gradient centrifugation.
E. coli ribosomal RNA was purified in a similar manner.
For the preparation of soluble RNA, the supernatant solu ion from the 80,000 x g centrifugation of the glass bead extract was utilized.
The initial SDS-phenol purification procedure was similar to that for ribosomal RNA.
Further purification was accomplished by the slow addition of solid sodium acetate to a final concentration of 3 M, with the resulting precipitate 1380 ATP-RNA Adenylyltransferase from P. putida Vol. 245,No. 6 being collected by centrifugation and discarded. Soluble RNA was then precipitated with ethanol, redissolved, and dialyzed extensively against 10 mu succinate-NaOH (pH 5.3). The resulting soluble RNA, which was free of detectable high molecular weight RNA as determined by sucrose density gradient centrifugation, was stored at -20".
Enzyme Assay,+The incubation temperature for all reactions was 37". The standard assay measured the conversion of 3H-ilTP into an acid-insoluble product.
Unless otherwise indicated, the incubation mixture (0.5 ml) contained 30 mM glycine-NaOH (pH 9.5), 0.8 mM aH-ATP, 20 mM MgCl,, saturating amounts of P. putida ribosomal RNA, and an appropriate amount of enzyme. Samples of 100 ~1 were withdrawn at various times and 5 ml of cold 10% trichloracetic acid were added to each sample. The acid-insoluble material was collected on membrane filters (Schleicher and Schuell, type B-6) which were then dried and monitored for radioactivity in a Packard Tri-Carb liquid scintillation spectrometer in a fluor of toluene containing 4 g per liter of 2,5-bis[2-(5-tert-butylbenzoxazolyl)]thiophene.
One unit of enzyme activity is defined as that amount of enzyme which catalyzes the incorporation of 1 nmole of AMP into an acid-insoluble product, with ribosomal RNA as primer, in 10 min at 37". Specific activity is expressed as units per mg of protein.
Reactions lacking either RNA, MgC12, or enzyme served as controls.
For the determination of inorganic pyrophosphate, the reaction mixture (0.5 ml) contained 30 mu glycine-NaOH (pH 9.5), 0.8 mu 3H-ATP, 20 mu magnesium acetate, 17 pg of P. putida ribosomal RNA, 0.2 mu UDPG, 0.2 mM NADP+, a suitable amount of enzyme and excess phosphoglucomutase, glucose 6-phosphate dehydrogenase, and UDPG pyrophosphorylase. The assay as described by Johnson et al. (42) couples the formation of inorganic pyrophosphate from ribonucleoside triphosphate polymerization to NADP+ reduction which is measured at 340 rnp in a spectrophotometer.
Calculation of the number of nmoles of NADPH formed was made with the molar extinction coefficient of 6.22 x lo3 (43).

Inorganic pyrophosphatase
was measured by the disappearance of inorganic pyrophosphate with the coupled UDPG pyrophosphorylase assay system. Adenylate kinase and ATPase were measured by the conversion of 3H-ADP and 3H-ATP, respectively, to other adenosine derivatives, which were separated by means of paper chromatography in isobutyric acid-iYH40H-HZ0 (66 : 1:33). Polynueleotide phosphorylase was measured under the same conditions as described for the standard assay except that 3H-ADP was used in place of 3H-ATP. (These conditions were found to be near optimal for the ADPincorporating activity isolated from the 80,000 X g pellet of P. putida.) Ribonuclease was assayed by measuring the loss of acid-insoluble radioactivity and the change in the sucrose density gradient profile of enzyme-treated l4C-ribosomal RNA.
Alkaline Hydrolysis of Polymeric Product-After incubation of the mixture for 60 mm, the reaction was terminated by the addition of 2 volumes of cold 5% HC104. The resultant precipitate was collected by centrifugation at 12,000 x g for 10 min and the supernatant solution was discarded. The pellet was washed once with 2 ml of cold 5% HC104 and twice with 2 ml of cold 1% HC104. After the addition of 2.0 ml of 0.3 N KOH, the dissolved pellet was incubated at 37" for 18 hours. The solution was then neutralized with Dowex 50 (H+), which was then removed by filtration through Whatman No. 42 filter paper and washed three times with 0.5 ml of 0.1 N NH40H. The combined filtrates were lyophilized, dissolved in HzO, and spotted with the appropriate standards on Whatman No. 3MM paper for electrophoresis and Whatman No. 41 paper for solvent chromatography.
Electrophoresis was carried out in 0.05 M ammonium formate (pH 3.6) at 36 volts per cm for 100 min. Descending solvent chromatography was performed in isobutyric acid-NH40H-HnO (66 : 1: 33). The radioactive lanes were cut out and counted in a liquid scintillation spectrometer.

Puri$cation of Enzyme
The entire purification procedure was performed at O-4". Preparation of Initial Extract-Frozen P. putida, 40 g of cells which had been grown to the early stationary phase, was homogenized in a Servall Omni-Mixer with 100 g of acid-washed glass beads and 25 ml of buffer (10 mM Tris-HCl (pH 8.0), 10 mu MgClz, 0.1 mu EDTA).
After 5 min of homogenization, 40 ml of the same buffer were added and the entire suspension was centrifuged at 12,000 x g for 10 min. The supernatant solution was decanted and recentrifuged at 12,000 x g for 10 min. The resulting supernatant solution was carefully decanted and diluted to 100 ml with the above buffer (glass bead extract).
Sedimentation-The glass bead extract enzyme fraction was centrifuged at 80,000 X g for 120 min. The supernatant solution, which was devoid of detectable ATP-incorporating activity, was poured off and discarded, and the pellet was stored at -20' overnight.
(No difference was observed if the purification was continued immediately without freezing the high speed pellet.) SoluInXzattin-The high speed pellet was thawed and homogenized in 100 mu glycine-NaOH (pH 10.6) by means of a glass tube fitted with a Teflon-tipped pestle. The resulting homogenate was centrifuged at 80,000 x g for 120 min and the supernatant solution was poured off and saved. The pellet was rehomogenized with 70 ml of the same buffer followed by centrifugation at 80,000 X g for 120 min. The supernatant solution was poured off and combined with the first (pH 10.6 supernatant, 134 ml).
pH Fractionation-Solid KC1 was slowly added with stirring to the pH 10.6 supernatant enzyme fraction to make the solution 1 M with respect to KCl. Hydrochloric acid (1 M) was then added dropwise with stirring until a pH of 5.0 was reached. A flocculent white precipitate formed during this procedure. After the suspension was stirred for 4 hours, the precipitate, which was devoid of ATP-incorporating activity, was removed by centrifugation at 12,000 X g for 10 min and discarded. The supernatant solution was dialyzed overnight against 4 liters of 10 mu Tris-HCl (pH 8.0)(pH 5 supernatant, 142 ml).
pH Concentration-The pH 5 supernatant enzyme fraction was adjusted to pH 3.2 by the slow addition of 1 M hydrochloric acid. A precipitate formed which, after stirring for 15 min, was collected by centrifugation at 12,000 X g for 10 min. The supernatant solution, which was devoid of ATP-incorporating activity, was poured off and discarded.
Xephadex G-100 Gel Filtration-Sephadex G-100, previously A 5-ml aliquot of the pH 3.2 precipitate enzyme fra&ion was applied to a col;mn, 43 X i.5 cm,-of Sephadex G"-100 eauilibrated with 10 mM Tris-HCl (DH 8.0) containing 1 M KCl. !&e column was developed at a flow'rate of0.5 ml per &in by the continual addition of the same buffer. Thirty milliliters of the 66-ml void volume were collected, followed by the 40 3-ml fractions presented in the figure. A---A, absorbance at 260 rnp; O---0, AMP incorporation into acid-insoluble product with ADP as the substrate; O--O, AMP incorporation into acidinsoluble product with ATP as the substrate.
Reaction mixtures (0.5 ml) contained 30 mM glycine-NaOH (pH 9.5), 20 mM M&12, 6.8 rn&xH-ATP (1.2 X lo3 cpm per nmole) or 3H-ADP (0.7 2 1Oi corn oer nmole). 84 UE of ribosomal RNA. and 50 ~1 of eluate fraction.' After in&bat& for 10 min, the reactionswere stopped by the addition of 5 ml of cold 10% trichloracetic acid and the amount of acid-insoluble product was determined as described under "Experimental Procedure ("Enzyme Assays") ." Fractions 24 through 29 were pooled and dialyzed against 10 mM Tris-HGl (pH 9.1)-(Sephadex G-100 enzyme fraction). equilibrated with 10 mM Tris-HCl (pH 8.0) containing 1 M KCl, was utilized to prepare a column, 43 x 2.5 cm. The pH 3.2 precipitate enzyme fraction was adjusted to 1 M with respect to KC1 by the addition of solid KCI. A 5-ml aliquot of the enzyme solution was then applied to the gel and the column was developed by the addition of 10 mM Tris-HCl (pH 8.0) containing 1 M KCI. The effluent pattern of the column is shown in Fig. 1. After washing the column with at least 100 ml of the high ionic strength buffer, another 5-ml aliquot of the enzyme solution was applied and eluted in the same manner.
This was repeated three more times for the remaining pH 3.2 precipitate enzyme fraction.
(In order to obtain satisfactory resolution of the applied material, it was important that the protein concentration of this fraction did not exceed 5 mg per ml.) The peak fractions from each column were pooled and dialyzed overnight against two successive 4-liter volumes of 10 mu Tris-HCI (pH 9.1)-(Sephadex G-100, 92 ml).
DEAE-Sephadex Chromatography-A column, 19 x 1.2 cm, of DEAE-Sephadex was prepared and washed extensively with 10 mM Tris-HCl (pH 9.1). The entire Sephadex G-100 enzyme fraction was applied to the column and subsequently eluted with a loo-ml linear gradient of 0 to 1 M KC1 in the same buffer. The enzyme eluted as a single peak at approximately 0.4 M KCl.
Cellulose Phosphate Chromatography-A column, 12.5 x 1.2 cm, of cellulose phosphate was prepared and washed extensively with 10 mM phosphate buffer (pH 7.1). The DEAE-Sephadex enzyme fraction was dialyzed overnight against the same phosphate buffer (DEAB-Sephadex (dialysis), 20 ml). As a result of this procedure, approximately two-thirds of the ATP incorporating activity was lost with the concomitant appearance of a very fine precipitate.
The entire suspension was applied to the cellulose phosphate column.
Elution was achieved with a loo-ml linear gradient of 0 to 1 M KC1 in 10 mM phosphate buffer (pH 7.1). The enzyme eluted as a single symmetrical peak at approximately 0.7 M KCl. The peak enzyme fractions were pooled (cellulose phosphate, 9 ml) and stored at 2".

Comments on Puri$cation Procedure
A summary of the purification procedure is presented in Table I. The data describe approximately a 360-fold purification with an apparent 5% recovery of initial enzyme activity. For the experiments described in this report, either the DEAE-Sephadex or the cellulose phosphate enzyme fractions were used.
Association with Macromolecules-An important aspect of the purification is the association of the enzyme with macromolecular cellular components.
Following the preparation of the glass bead extract and removal of cellular debris, high speed centrifugation of the enzyme extract was performed.
Although most cellular enzymes, including DNA-dependent RNA polymerase, remain in the supernatant solution during this procedure, the ATP-incorporating activity was detected only in the pellet. Solubilization of the enzyme was accomplished by homogenization of the high speed pellet in alkaline buffer, pH 10.6. At pH 10.1 only about 50% of the enzyme was released into the supernatant fraction, while less than 10% was solubilized at pH 9.8. Even after solubilization from the high speed pellet, the enzyme was found to aggregate with other cellular components. The pH 5 fractionation and Sephadex G-100 gel filtration steps were ineffective unless carried out in the presence of 1 M salt.
Removal of Contaminating Nucleic Acid-Even though the majority of the Aneo-absorbing material was separated from the ATP-incorporating activity in the gel filtration step (Fig. l), the Azso:A260 ratio of the enzyme fractions remained at 0.5. At this point in the purification, the residual nucleic acid was still in sufficient quantity to prevent any stimulation of the ATPincorporating activity by the addition of exogenous RNA. The next step, DEAE-Sephadex chromatography, was utilized to remove most of the remaining Azso-absorbing material.
The Azso:A260 ratio of the eluate fractions in the region of the enzyme was approximately 1.0. Following this step, the ATP- Stability of Enzyme-The stability of the enzyme fractions, up to and including the DEAE-Sephadex fraction, was such that, when stored at 2", no loss in activity could be detected for at least 1 month.
However, dialysis of the DEAE-Sephadex enzyme fraction led to a considerable loss in activity.
The cellulose phosphate enzyme fraction was also unstable, losing about 50% of its activity in 1 week.
Contaminating Act&ties-When examined at the concentra-

Characteristics of Reaction
General Properties-Both the rate and extent of ribonucleotide incorporation into an acid-insoluble product increased with temperature and were optimal at 37". Higher temperatures, up to 45", produced an equally efficient initial reaction but the total incorporation was significantly inhibited. The optimal pH for the reaction was 9.5 with either 30 mM glycine-NaOH buffer or Tris adjusted to that pH with HCl. Ribonucleotide incorporation equal to approximately 50% of the optimal activity was observed with Tris-HCl buffer at pH 8.5 and with glycine-NaOH buffer at pH 8.  pg of the cellulose phosphate enzyme fraction to the standard reaction mixture resulted in the incorporation of 0.41, 1.8, and 3.8 nmoles of AMP, respectively, in 20 min. Similar proportionality was obtained with the less purified fractions.
The kinetics of the reaction was consistently biphasic (Fig. 2), independent of the purity of the enzyme or the pH of the reaction mixture.
All of the studies presented in this report were Metal Ion Requirement-The enzyme requires the presence of magnesium ion for activity (Fig. 3). There was no detectable activity when Mgzf was omitted from the assay. The activity increased with increasing M, &-until a maximum between 15 and 30 mM, followed by considerable inhibition between 30 and 40 mM.
The apparent K, for Mg2+ as determined from the Lineweaver-Burk plot shown in the inset of Fig. 3 (32) have reported an ATP-incorporating activity in E. coli which requires both Mg2+ (25 mM) and M$+ (2 mM) for optimal activity.
In the presence of Mg2+, MI?+ produced approximately a Y-fold stimulation of the activity. However MnZ+, in the presence of optimal Mg"+, did not stimulate the ictivity of the P. putida enzyme but instead inhibited the process. Fifty per cent inhibition was accomplished by the addition of about 4 mM Mn 2+ to the reaction mixture, while the addition of an equivalent amount of Mg2+, giving a final concentration of 24 mM Mg2+, still produced near maximal activity (see Fig. 3). A similar inhibitory effect of Mnzf was observed at pH 8.
Ribonucleoside Triphosphate Requirement-Of the four common ribonucleoside triphosphates, only ATP was significantly incorporated into a polymeric product by the enzyme (Table  II).
Under identical conditions, the rate of incorporation of CMP was about 7% that of AMP, whereas the incorporation of UMP or GMP was negligible (less than 1%). If either GTP, UTP, or CTP was added to the reaction mixture in addition to ATP, incorporation of AMP was inhibited approximately 30%. The mechanism of this inhibition is unknown. Neither ADP (see Fig. 2) nor d-ATP (data not shown) could replace ATP as a substrate in this reaction.
The response of the enzyme to increasing ATP concentration is presented in Fig. 4. The apparent K, for ATP as determined from the Lineweaver-Burk plot shown in the inset of Fig. 4 was 3 X 1OW M.
RNA Requirement-Incorporation of AMP into a polymeric product by the enzyme was dependent upon exogenous RNA. In the absence of added RNA, no acid-insoluble product was detected even after incubation for 3 hours (see Fig. 2). Ribosomal RNA, relatively independent of its source, soluble RNA, and poly C were capable of fulfilling this requirement (  Samples were withdrawn at the times indicated and product formation was determined as described under "Experimental Procedure ("Enzyme Assays") ." III), while poly A, poly I, poly U, DNA, and (pT)s were minimally effective (less than 10% of the rate with ribosomal RNA).
The enzyme system approached saturation with P. putida ribosomal RNA at approximately 40 pg per ml (Fig. 5). Although the maximal observed velocities for both ribosomal and soluble RNA were nearly identical, saturation with soluble RNA was not obtained until about 140 pg per ml (Fig. 6). Assuming a molecular weight of 2.5 x lo4 daltons, the apparent K, for soluble RNA as determined from the Lineweaver-Burk plot shown in the inset of Fig. 6 was 1 X 10e6 M. A similar plot for ribosomal RNA (l/v plotted against l/rRNA) did not yield a straight line (Inset a of Fig. 5). A Hill plot (45,46) of these data yielded a straight line with a slope of 1.8. A plot of l/v against l/rRNAlJ is presented in Inset 6 of Fig. 5. Assuming an average molecular weight of 8.2 X lo5 daltons for the two species of ribosomal RNA, the apparent K, was 6 x lop9 M.
The oligoribonucleotide, ApApApA, could also function in this reaction.
At the saturating concentration of 50 pg per ml, ApApApA, after a lag of about, 20 min, was capable of stimulating the incorporation of AMP into acid-insoluble product (Fig.  7). The rate was approximately 50% of that observed with ribosomal RNA.
Shorter oligoribonucleotides of adenylic acid, ApApA or ApA, did not function in this reaction.
The oligoribonucleotide, ApApApAp, which lacks the free 3'-OH due to the presence of a phosphate group, was also ineffective. When added to the standard reaction mixture which contained ribosomal RNA, ApApApAp inhibited the incorporation of AMP by approximately 50%, whereas the addition of an equivalent, amount of ApApApA to the standard reaction mixture had no inhibitory effect.

Characterization of Reaction Products
Xtoichiometry of Reaction Products-As measured by means of the production of NADPH in the coupled assay with UDPG pyrophosphorylase (see "Experimental Procedure ("Enzyme Assays")"), inorganic pyrophosphate was shown to be a product of the reaction.
The amount of inorganic pyrophosphate produced was equivalent to the amount of AMP incorporated into acid-insoluble product.
In two separate experiments, when 11.4 and 8.2 nmoles of inorganic pyrophosphate were produced in 40 min, 10.8 and 7.4 nmoles of AMP were incorporated into polymeric product, respectively.
When ribosomal RNA was omitted from the reaction mixture, neither acid-insoluble product nor inorganic pyrophosphate was formed.

Chain
Length Determination of Polymeric Product-Alkaline hydrolysis of the acid-insoluble product formed when aH-ATP was used as substrate in the standard reaction mixture, followed by paper chromatography of the hydrolysate in isobutyric acid-NH40H-H20, gave the results shown in Table IV Nearest neighbor analysis of polymeric product The product was prepared in a reaction mixture (1.0 ml) containing 30 rnM glycine-NaOH (PH 9.5), 0.8 mM c+P-ATP, 20 mM MgC&, 33.6 fig of ribosomal RNA, and 30 pg of the DEAE-Sephadex enzyme fraction.
The reaction was terminated at 30 min by the addition of 2 ml of cold 5y0 HClOd and the polymeric product was isolated and hydrolyzed as described under "Experimental Procedure ("Alkaline Hydrolysis of Polymeric Product")." An aliquot of the alkaline hydrolysate was subjected to electrophoresis and the paper was analyzed for radioactivity. to 96% of the radioactivity cochromatographed with 2'(3')-AMP (RF = 0.66), while 0.5 to 0.7% moved with adenosine (Rr = 0.83). The remaining radioactivity chromatographed as a single peak (unknown) with an RF of 0.46. Upon electrophoresis of the alkaline hydrolysate, this material migrated slightly ahead of 2'(3')-AMP and was well separated from the adenosine di-, tri-, and tetraphosphate regions.
Since it is known that, under the conditions used, the hydrolysis of poly A may be only 95% complete (47), this radioactive material most probably represented unhydrolyzed oligomer(s) of adenylic acid.
The length of the average polymeric product can be approximated by dividing the amount of 2'(3')-AMP, which represents the internal residues of the chain, by the amount of adenosine, which represents the 3'-external residue.
The results presented in Table IV indicate a length in the range of 100 to 200 adenylate residues per chain. Incomplete hydrolysis of the product and the slow conversion of 2'(3')-AMP to adenosine (48) imply that this approximation is a minimum value. The absence of any radioactive material which might represent a 5'-external residue suggests that the adenylate chain probably was not synthesized de novo.
Nearest Neighbor Analysis of Polymeric Product--When LU-~~P-ATP was used as substrate and the resultant product was subjected to alkaline hydrolysis and analysis by paper electrophoresis, 94% of the radioactivity migrated with 2'(3')-AMP (Table  V). A small amount of radioactivity (about 3a/,, data not shown in Table V) migrated slightly ahead of 2'(3')-AMP, corresponding to the unhydrolyzed oligomer(s) of adenylic acid observed in the chain length analysis.
The remainder of the radioactivity was divided among the other three common 2'(3')ribonucleoside monophosphates. Some of the radioactivity in the 2'(3')-CMP region undoubtedly resulted from trailing of 2'(3')-AMP which migrates just in front of 2'(3')-CMP in this system. These data, along with the results of the chain length analysis, are consistent with a polymeric product composed of long chains of adenylate residues which are attached to the 3'-end of the added ribosomal RNA.
Sucrose Density Gradient Analysis of Polymeric Product-Reaction products, prepared in standard assay mixtures containing ribosomal RNA and 3H-ATP, were analyzed by SDS-sucrose density gradient centrifugation as described in the legend to 1 t 12 24 36 FRACTION NUMBER FIG. 8. Sucrose density gradient analysis of polymeric product. Reaction mixtures (0.25 ml) contained 30 mM Tris-HCl (pH 8.5), 0.8 mM 3H-ATP (4.5 X lo3 cpm per nmole), 20 m&c MgC12, 1.5 pg of the cellulose phosphate enzyme fraction, and P. putida ribosomal RNA, 33.6 pg in A and 336 pg in B. After incubation for 19 min, 302 pg of ribosomal RNA were added to (A) and 119 pg of soluble RNA were added to each. At 20 min the reaction was stopped by cooling in ice and the addition of 30 ~1 of 1% SDS. Samples of 100 ~1 were withdrawn from each and treated as described for the radioactive assay under "Experimental Procedure ("Enzyme Assays")." Another 100 ~1 were withdrawn from each and layered on separate sucrose gradients (5 to 20yG) in 50 mM Tris-HCl (pH 8.1) with 0.1% SDS. The gradients were centrifuged at 22" for 53 hours at 39,000 rpm in a Spinco SW 39 rotor. The tubes were punctured and 46 fractions were collected. The odd-numbered fractions were measured for absorbance at 260 mp, following the addition of 0.4 ml of Hz0 to each, and the even-numbered fractions were used to determine the acid-insoluble radioactivity. Recovery of radioactivity was 87y0 in A and 927, in B. Fig. 8. The data presented in this figure show that the major portion of the product sedimented in the region of the 23 and 16 S ribosomal RNA species. Chains of adenylate residues, 100 to 200 nucleotides in length, would not be sufficient to produce these large sedimentation values. Since noncovalent binding of poly A to ribosomal RNA does not occur even in the presence of divalent cations (49,50), it was concluded that the polymeric product was covalently attached to the ribosomal RNA.
These results also show that both species of ribosomal RNA can fulfill the requirement for exogenous RNA in this reaction. Taking into account that 23 S ribosomal RNA has twice the molecular weight of 16 S ribosomal RNA, the amount of radioactive product associated with the smaller species is slightly greater (1.3 to 1.6 times) than that associated with the larger species.

DISCUSSION
The ATP-incorporating activity described in this report, which was isolated from the 80,000 X g pellet of P. putida, was purified from major contaminating activities which might interfere in the assay. These activities included ATPase, adenylate kinase, polynucleotide phosphorylase, RNase, and inorganic pyrophosphatase.
Furthermore, the purified enzyme was essentially free of endogenous RNA.
In the presence of magnesium ion and exogenous RNA, the enzyme specifically cata-