REVERSIBLE SYNTHESIS OF POLYRIBONUCLEOTIDES WITH AN ENZYME FROM ESCHERICHIA COLI* BY URIEL 2. LITTAUERt AND ARTHUR KORNBERG (Frm the Department of Microbiology, Washington University School of Medicine, St. Louis, Missouri) (Received for publication, October 15, 1956) Studies on nucleotide and coenzyme synthesis in this laboratory led to an attempt to convert nucleoside 5'-polyphosphates to polyribonucleo- tides. We were able to show that extracts of Escherichia coli convert C14-adenine-labeled ATP1 to an acid-insoluble nucleotide, and that the addition of adenylate kinase increased the rate of the reaction (1, 2). The discovery of Ochoa and coworkers (3, 4) that polyribonucleotides are formed from nucleoside diphosphates in a reaction catalyzed by an enzyme from Azotobacter vinelandii made it clear that ADP rather than ATP was the reactive component. In preliminary reports (1, 2) we have described briefly an enzyme from E'. coli which converts nucleoside diphosphates to polynucleot,ides according to thc following general equation : n nucleoside-PP d (nucleoside-P), + nP, A similar enzyme was found in an extract from Micrococcus Zysodeikticus by Beers (5). The purpose of this report is to describe the purification of the E. coli enzyme, its properties, and the stoichiometry of the reversible phosphorolysis which it catalyzes. Materials 5-Phosphoribosyl pyrophosphate was prepared from ribose 5-phosphate and ATP with a pigeon liver enzyme (6). CI4-A5P was prepared from adenine-8-CI4 (Isotopes Specialties Company, Inc.) and 5-phosphoribosyl pyrophosphate with yeast A5P pyrophosphorylase (7). C14-ATP was prepared from the labeled A5P by the combined action of yeast adenylate * This investigation was aided by grants from the National Institutes of Health, Public Health Service, and the National Science Foundation. t Fellow of the Dazian Foundation for Medical Research. Permanent address, The Weizmann Institute of Science, Rehovoth, Israel. 1 The abbreviations used are adenosine 5'-phosphate, A5P; adenosine diphosphate, ADP; adenosine triphosphate, ATP; cytidine diphosphate, CDP; deoxyribonuclease, DNAase; deoxyribonucleic acid, DNA; guanosine diphosphate, GDP; inorganic orthophosphate, l', ; P32-lnbeled Pi, P,32; ribonuclease, RNAase; ribonucleic acid, RNA; thymidine diphosphate, TDP; tobacco mosaic virus, TMV; turnip yellow mo- saic virus, TYMV; tris(hydroxymethyl)aminomethane, Tris; uridine 5'-phosphate, U5P; uridine diphosphate, UDP. 1077 1078 SYNTHESIS OF POLYRIBONUCLEOTIDES kinase (8) and pyruvate phosphokinase with phosphopyruvate, and was purified on a Dowex 1 column (9). C14-ADP was prepared from C14-ATP and glucose by the action of purified yeast hexokinase and isolated by chromatography on a Dowex 1 column (9). Labeled U5P was prepared from ~racil-2-C~~ (Isotopes Specialties Company) and 5-phosphoribosyl pyrophosphate with U5P pyrophosphorylase obtained from L. bz$dus.2 C14-UDP was prepared from labeled U5P by the action 01 yeast iiucleoside monophosphate kinase (8) and pyruvate phospholunase with phospho- pyruvate. UDP was separated from U5P and UTP by Dowex 1 col- umn chromatography (8). Thymidine diphosphate was prepared by the method of Hall and Khorana as described for the uridine nucleotides (10). Unlabeled ADP, UDP, CDP, and GDP were obtained from the Sigma Chemical Company. Protamine sulfate was generously supplied by Eli Lilly and Com- pany. Purified potato starch was obtained from the Fisher Scientific Company. Crystalline RNAase and DNAase were Worthington Biochem- ical Corporation products. Venom phosphodiesterase free from mono- esterase was generously given to us by Dr. L. A. Heppel and Dr. L. Shuster Methods A5P was assayed with Schmidt's deaminase by Kalckar's method (11); ADP and ATP were determined enzymatically as previously described (12, 13) ; U5P7 UDP, and UTP were separated by ion exchange chromatog- raphy and determined spectrophotometrically (8). Orthophosphate was estimated by the method of Fiske and Subbarow (14) ; acid-labile phos- phate was the orthophosphate liberated after a 10 minute hydrolysis in 1 N HCl at loo", and total phosphate was measured as orthophosphate after being ashed with a sulfuric-nitric acid mixture. Pentose was deter- mined by the Mejbaum procedure (15), with a 45 minute heating period and A5P as a standard. Proteins were determined by the phenol method of Lowry et al. (16). Ion exchange chromatography wm carried out at 2' with an automatic fraction collector on Dowex 1 columns (2 per cent cross-linked, 200 to 400 mesh, chloride form) (9). P32 was measured as a thin, dried layer on metal disks under a Geiger-Muller tube. CI4-con- taining samples were plated as thin layers on metal disks and measured in a gas flow counter. Self-absorption corrections were applied as indicated. Enzyme A ssays-The enzyme activity was determined in three ways. Assay A: Incorporation of Labeled Nucleoside Diphosphate into Acid- Insoluble Precipihte-The incubation mixture (0.25 ml.) contained 0.05 ml. of glycylglycine buffer (1 M, pH 7.4), 0.02 ml. of ADP (0.04 M), 0.02 ml. of 8-CI4-ADP (0.00227 M, 7.8 X lo6 c.p.m. per pmole), 0.01 ml. of * See Crawford, et al. (22). IJ. Z. LITTAUER AND A. KORNBNItG 1079 MgClz (0.1 M), and less than 0.3 unit of enzyme. A similar procedure was used for studying CI4-UDP incorporation. After incubation at 37" for 10 minutes, the reaction was stopped by immersing the tubes in an ice bath; 0.5 ml. of carrier nucleic acid in the form of a 1:20 dilution of crude E. coli extract and 0.25 ml. of 7 per cent perchloric acid were then added. After 10 minutes in the cold, the precipitate was centrifuged, washed twice with 1.0 mi. portions of 1 per cent perchloric acid, and once with 1.0 ml. of 0.01 N HC1. The precipitate was dissolved in 0.4 ml. of 0.05 M KOH. A 0.10 ml. aliquot was removed to a planchet, dried, and assayed for radio- activity (self-absorption correction factor, 1.55). 1 unit of enzyme was defined as the amount catalyzing the incorporation of 1.0 pmole of ADP in 1 hour, and the specific activity was expressed as units per mg. of pro- tein. Under these assay conditions, the radioactivity in the acid-insoluble precipitate was proportional to the enzyme concentration. Thus, with use of 0.01,0.03,0.06, and 0.12 ml. of a crude enzyme fraction, 0.005,0.017, 0,027, and 0.048 pmole of ADP, respectively, were incorporated in the polynucleotide. Assay B: 8-C14-ATP Incorporation in Presence of Myokinase-The incu- bation mixture (0.25 ml.) contained 0.02 ml. of Tris buffer (1 M, pH 8.0), 0.02 ml. of ATP (0.05 M), 0.01 ml. of 8-CI4-ATP (0.0087 M, 3.8 X lo6 c.p.m. per pmole), 0.04 ml. of ADP (0.006 M), 0.01 ml. of MgCL (0.1 M), 0.02 mi. of yeast adenylate kinase (heated ethanol fraction (8)), and less than 0.25 unit of enzyme. The mixture was incubated for 10 minutes at 37". The reaction was stopped and the precipitate treated as in Assay A. An enzyme unit was defined as in Assay A, and equally good proportion- ality of the values obtained to the amounts of enzyme added was observed. Assay C: Nucleoside Diphosphate Exchange with P3*-This assay was based on that of Grunberg-Manago et al. (4). The incubation mixture (0.5 ml.) contained 0.10 ml. of glycylglycine buffer (1 M, pH 7.4), 0.10 ml. of nucleoside diphosphate (0.004 M), 0.05 ml. of in potassium phos- phate buffer, pH 7.4 (0.0052 M, 5.2 X lo6 c.p.m. per pmole), 0.02 ml. of MgCla (0.1 M), and less than 0.3 unit of enzyme. The mixture was incu- bated for 20 minutes at 37". The reaction was stopped by immersing the tubes in an ice bath, adding 0.5 ml. of 5 per cent perchloric acid to acidify the solution, and 0.10 ml. of an acid-washed Norit A suspension (10 per cent dry weight) to adsorb the nucleotides. After 10 minutes in the cold, the Norit was centrifuged and washed three times with 2.5 ml. portions of water. The precipitate was suspended in 0.8 mI. of 50 per cent ethanol containing 3 ml. of concentrttted NEIrOH per liter. An aliquot (0.2 ml.) of the ttbovc Norit suspension was (fried on a planchet arid the radioactiv- ity measured (self-absorption correction factor 1.15). 1 unit of enzyme was defined as the amount causing the incorporation of 1.0 pmole of P3* 1080 SYNTHESIS OF POLYRIBONUCLEOTIDES into ADP per hour, and the specific activity was expressed as units per mg. of protein. The amount of Pi incorporated into the terminal phos- phate of ADP was calculated from the equation: total 0.p.m. in ADP initial specific activity of Pi pmoles phosphate incorporated = These values are somewhat low since no correction was made for the de- crease in specific radioactivity of the Pi by the contribution of the ter- minal phosphate of ADP. However, since the initial rates were deter- mined when the exchange was still less than 10 per cent of completion, this correction would be very small. With 0.01, 0.03, 0.06, and 0.08 ml. of enzyme (manganese supernatant fluid), 0.0028, 0.011, 0.021, and 0.028 pmoles of the Pi were found to be incorporated into the ADP. Growth of Cells and Preparation of Cell-Free Extracts-E. coli strain B was grown in a medium (pH 6.8 to 7.0) containing 1 per cent yeast extract (Difco, dehydrated), 1 per cent glucose, 2.18 per cent K2HPO4, 1.70 per cent KH2P04, and about 20 mg. per liter of Antifoam A (Dow-Corning); glucose was autoclaved separately and added to the cooled medium. 15 liters of the medium in a 20 liter Pyrex bottle were inoculated with 1.5 liters of a 14 to 16 hour culture and incubated at 37" with vigorous forced aeration until the end of the logarithmic phase of growth (3 to 4 hours). The cells were harvested in a Shnrples supercentrifuge (8 gm. of wet cells per liter) and washed with 4 volirmes of cold 0.9 per ccnt KCl. The cells (170 gm., wet weight) were suspended in 0.05 M glycylglycine buffer, pH 7.4, to a final volume of 680 ml. and placed for 10 minutes in a Raytheon 10-kc. oscillator at 6-8". The residue was collected by centrifugation for 20 minutes at 10,000 X g in a Serval1 centrifuge and the supernatant Auid (10 minute sonic extract) was discarded. The residue was suspended in the same buffer at a final volume of 680 ml., subjected again to sonic oscillation for 30 minutes, and centrifuged for 20 minutes 8.5 before, The turbid supernatant fluid (sonic extract of the residue) was used for fur- ther purification steps. The cells were treated in the oscillator immedi- ately after harvesting in order to obtain most of the enzyme activity in the residue of the 10 minute sonic extract. This residue, as well as the extract of it, was stored at -15" for over 2 months without loss of enzyme activity. Results Purification of Enzyme All operations were carried out at 0-3" except as indicated. Mn++ and Protamine Steps-To 615 ml. of the sonic extract of the resi- due (Table I), 31 ml. of 1 M NInClz were added slowly with mechanical U. Z. LITTAUER AND A. KORNBERG 1081 activity Specific stirring; the stirring was continued for 30 minutes. Insoluble material was removed by centrifugation for 20 minutes at 10,000 X g (manganese supernatant fluid). (For assays at this stage this fraction must be dia- lyzed overnight against 0.9 per cent KCI.) To the undialyzed, clear yel- low supernatant fluid, 61 ml. of l per cent protamine sulfate were added with mechanical stirring during a period of 10 minutes. (The amount of protamine sulfate needed to precipitate over 90 per cent of the activity was determined for each run.) The precipitate formed was collected by centrifugation, suspended in 200 nil. of 0.05 M potassium phosphate buffer, pH 7.5, and recentrifuged. The supernatant fluid mas dialyzed overnight against 6 liters of 0.9 per cent KCl and became slightly turbid (protamine eluate). This fraction was stored at - 15" for no longer than 1 week. 280:X260 TABLE I Purification of Enzyme units per mg. protein 0.1 0.5 11 17 28 Step 0.55 0.62 1.75 1.75 1.75 Units per ml.* 10 min. sonic extract. ................. Sonic extract of residue.. ............. Protamine eluate.. .................... Ethanol I.. ........................... 11.. it .......................... 1.1 3.0 9.8 13.1 4.8 Tqtal units* 670 1860 1960 870 320 Ethanol Fractionation-To 200 ml. of the protam Protein mg. ficr ml. 10.9 6.9 0.88 0.76 0.17 le eluate fraction were added 4.0 ml. of 1 M potassium acetate (pH 5.5) and then 1.2 ml. of 0.5 M ZnCla (adjusted to pH 5.5 with acetic acid). After standing for 10 min- utes, any precipitate which formed was removed by centrifugation for 3 minutes at 10,000 X 9. To the supernatant fluid, 44 ml. of 50 per cent ethanol (-15") were added over a 7 minute interval, during which time the mixture mas chilled to -2". The precipitate was removed by centrifu- gation for 3 minutes at 10,000 X g, and to the supernatant fluid 52 ml. of 50 per cent ethanol were added as described above, the temperature being maintained at -2" to -4". The precipitate was collected by centrifuga- tion and dissolved with 0.05 M Tris buffer, pH 8.0, to a final volume of 67 ml. (Ethanol I). To 66 ml. of the Ethanol I fraction were added 0.5 N acetic acid to pH 5.5 (approximately 2.5 ml.) and then 0.40 ml. of 0.5 M ZnC18. After 10 minutes the precipitate was centrifuged as before, and to the supernatant fluid 13.5 ml. of 50 per cent ethanol (- 15") were added during a 7 minute period; the mixture was chiIled to -2" during this interval. The precipi- 1082 SYNTHESIS OF POLYRIBONUCLEOTIDES tate was centrifuged and dissolved in GG ml. of 0.05 M Tris buffer, pII 8.0 (Ethanol 11). The optical density as indicated by the X280:X260 ratio varied for cliffercnt enzyme preparations from 1.5 to 1.75. Purification of the Ethanol I fraction could also be achieved by starch column electrophorcsis. The column (3.5 X 50 em.) was washed first with water and then with 0.05 M Tris buffer, pH 8.0. Ethanol I fraction (4.0 ml.) was added and the column was washed twice with 5.0 ml. of the Tris buffer. Current was applied for 19 hours (500 volts, 8 to 10 ma.). The enzyme was eluted with 0.05 M Tris buffer, pH 8.0 (8.0 ml. per hour), and 4.0 ml. fractions were collected. The enzyme appeared between Frac- tions 19 and 27 with an over-all recovery of 83 per cent. Fractions 20 and 21 contained 18 per cent of the total activity with a 10-fold increase in specific activity over the Ethanol I fraction; $he enzyme was very labile in this state and activity was lost rapidly upon freezing and thawing. Stability of Enzyme-A purified enzyme fraction (Ethanol 11) retained about 70 per cent of its activity after storage for 2 months at -lo", but, when diluted (1:lO in 0.10 M glycylgIycine buffer, pH 7.4), the activity was rapidly lost. Heat inactivation of the enzyme (manganese super- natant fluid in 0.10 M glycylglycine buffcr, pH 7.4) was observed by heat- ing for 5 minutes at 60" or 70"; 70 and 98 per cent, respectively, of the original activity was lost. Presence of Other Enzymcs-DNAase activity was not detected in the purified enzyme (0.08 unit of Ethanol I1 degraded less than 0.10 per cent of P32-labeled Te phage DNA (0.25 y of DNA, 1.7 X IO4 c.p.m.) after a 20 minute incubation period at 37". Adenylate kinase activity was low (1.0 unit of Ethanol I catalyzed the Formation of 0.09 pmole of ADP per hour from ATP and A5P when measured with the coupled pyruvate phos- phokinase-lactic dehydrogenase system (13)). An amount of RNAase activity was present in 1.0 unit of enzyme (Ethanol I) sufficient to liberate 0.05 pmole of mononucleotide per hour when incubated with yeast sodium nucleate in the absence of phosphate buffer. The crude extract contained a nucleotide-N-ribosidase3 which hydrolyzed A5P (and thus in the pres- ence of adenylate kinase removed ADP from the reaction). However, this activity was absent in the purified fractions; 1.5 units of the Ethanol I fraction liberated less than 0.002 pmole of ribose 5-phosphate per hour. Incorporation of Nucleoside Diphosphates into Polynucleotides Balance Study of Reaction-For each micromole of acid-labile phosphate and ADP disappearing, 1 pmole of Pi was liberated, and approximately equivalent amounts of pentose and phosphate appeared in the acid-insolu- 3 We are grateful to Dr. J. Hurwitz, Dr. L. A. Heppel, and Dr. €3. I,. Horecker for informing us of their unpublished work on this enzyme. U. Z. LITTAUER AND A. KORNBERQ 1083 ble product (Table 11). The isolated polyadenylate was hydrolyzed with 1 N KOH for 15 hours at 37", and the quantity of adenosine 2'- and 3'-phos- phates found matched the disappearance of an equivalent amount of ADP, TABL~ I1 Balance Study of ADP, UDP, CDP, and GDP Incorporation into Polynucleotides f0.42 -0.38 -0.36 52 - Orthophosphate, A pmole.. ... Total phosphate, A pmole .... Pentose, A pmole.. ........... Acid-labile phosphate, A pmolt Ultraviolet density, A pmole. . Radioactivity, % incorpora- tion, ..................... Acid-labile phosphate) % de- crease .................... $0.21 -0.21 +O. 41 +0.39 -0.25 +0.28 +Os 36 57 30 32 ADP (1) 1 UDP (2) Acid- 1 1 Acid- I soluble soluble Po'ymer CDP (3) I GDP (4) (1) At zero time, the Pi, acid-labile phosphate, and "adenosine" values were 0.07,0.75, and 0.86 pmoles, respectively; after 60 minutes, the respective values were 0.49, 0.37, and 0.50 pmoles. (2) At zero time, the Pi, acid-labile phosphate, and "uridine" values were 0.10, 0.64, and 0.75 pmoles, respectively; after 60 minutes the respective values were 0.31, 0.43, and 0.50 pmoles. (3) At zero time, the Pi, acid- labile phosphate, and "cytidine" values were 0.06, 0.74, 0.87 pmoles, respectively; after 60 minutes the respective values were 0.50, 0.30, and 0.43 pmoles. (4) At zero time, Pi, acid-labile phosphate, and guanine values were 0.11, 0.64, and 0.80 pmoIes, respectively; after 60 minutes the respective values were 0.10, 0.66, and 0.79 umoles. The reaction mixture (0.25 ml.) was as described in Assay A (see the text), with 0.8 unit of Ethanol I and an incubation period of 60 minutes at 37". In the ADP experiment, 0.04 ml. of ADP (8-C14, 0.00147 M, 6.17 X lo6 c.p.m. per pmole) was used; in the UDP experiment, 0.02 ml. of UDP (2-Cl4, 0.00522 M, 1.22 X 106 c.p.m. per pmole) was used; in the CDP and GDP experiments, the substrates were not labeled. The reaction was stopped by immersing the tubes in an ice bath, adding 0.5 ml. of a cold solution of crystalline serum albumin (1.6 mg. per ml.) and 0.25 ml. of 7 per cent perchloric acid. After 10 minutes in the cold, the precipitate was removed by cen- trifugation. The supernatant fluid is the "acid-soluble" fraction. The precipi- tate, washed twice with 1.0 ml. portions of 1 per cent perchloric acid and once with 1.0 ml. of 0.01 N HCl, and dissolved in 0.4 ml. of 0.025 N KOH, is the "polymer" fraction. The optical density of the polymer fraction was determined after hy- drolysis of an aliquot in 1 N KOH for 15 hours. The fraction of the total radioactivity found in the product derived from the added C14-AI>P (57 per cent) was in agreement with the fraction of the acid-labile phosphate disappearing (52 per cent) . Similar results were obtained in balance studies carried out with C14- UDP and with CDP, but with GDP no reaction was detected (Table 11). 1084 SYNTHESIS OF POLYRIBONUCLEOTIDES However, in GDP experiments with large amounts of enzyme (5.0 units) and a longer incubation time (3 hours), an acid-insoluble product was ob- tained which, upon perchloric acid hydrolysis, yielded guanine, as deter- mined by paper chromatography. When the reaction was carried out in the presence of all four nucleoside diphosphates, the extent of GDP in- corporation increased considerably, and approximately equivalent amounts of each of the four nucleotides were found in the polymer. Extent of Reaction-In the crude extract only small amounts of product accumulated, and when the incubation was continued for longer periods the product disappeared (Fig. 1). On the other hand, when a purer en- MI NU T ES FIG. 1. The extent of ADP incorporation as a function of time. The reaction mixtures (0.25 ml.) were as described in Assay A. The incubation temperature wa6 at 37" for the time indicated. The crude enzyme was 0.18 unit of a 30 minute sonic extract and the purified enzyme was 0.72 unit of an Ethanol I1 fraction. zyme fraction (Ethanol I) was used, more than 50 per cent of the ADP was converted into polyadenylate, and this did not disappear on continued incubation. With sufficient amounts of enzyme and longer incubation periods, extensive polymerization of even low concentrations of ADP (Le. 10-4 M) was observed. Efect of Substrate and Mg++ Concentratim-The K, value for ADP was found to be 2.0 X At this concentration of ADP, 1 mg. of the Ethanol 11 fraction polymerized 200 pmoles of ADP per hour. The reac- tion was found to require Mg++. When this was omitted, no incorpora- tion of ADP could be detected. Optimal Mg++ levels depended on the ADP concentration and were generally attained at an ADP: Mg++ ratio of 1.5; inhibit.ing effects were observed with lower ratios. At an ADP: Mgf+ ratio of 0.475, the rate n.as only 46 per cent of t.he maxiriiltl rate. Isolation os Pro~ucts-PolyadeIlylatc was isolated from an iiicubatioii M. mixture similar to that described for Assay A, with use of 0.016 M ADP (8-C14) (2.7 X lo4 c.p.m. per pmole) and 0.5 unit of enzyme. After 60 minutes at 37", 0.20 ml. of 1.0 M sodium acetate buffer, pH i3.5, was added, and, after 30 minutes in the cold, a transparent, gel-like precipitate ap- peared. The precipitate was centrifuged for 10 minutes at 10,000 X g, washed twice with 1.0 ml. portions of cold water; and dissolved in 0.25 ml. of 1.0 M Tris bufl'cr, pH 8.0. This gave a highly viscous solution from which threads could be drawl. The solution lost its viscosity when stored in the cold for 23 hours. When polyadenylate was precipitated with a stronger a.cid, such as 1.7 per cent perchloric acid, the precipitate obtained failed to give a viscous solution. The absorption spectrum of the polymer was different from that of AMY; the X250:h260 and h280:X260 absorption ratios (pH 8.0) were 0.93 and 0.32, respectively. When incubated with 1 x KOH for 15 hours at 37", the polymer was rendered completely acid- soluble and the absorption spectrum was identical to that of adenosine :3'-phosphatc. MgCh (0.1 M) precipitated polyadcnylate, and thc pre- cipitate could he redissolved with 1.0 M Tris buffer, pH 8.0. In the case of UDP and CDl', the polymerized products did not precipitate at pH 3.5, and with the use of stronger acid (I .7 per wilt perchloric acid) t,ho rwov- eries were poor. -4lthough all the polymers wcre precipitated readily in the preseim of' 80 per cent alcohol, significant amounts of the nucleoside diphosphates were coprecipitated. l'recipitation with streptomycin (17) provided the most satisfactory procedure : sodium acetate buffer (0.05 ml., 1 M, pH 3.5) and streptomycin sulfate (0.1 ml. of a 10 per cent solution) were added to t.he reaction mixture. After 10 minutes at Oo, the precipi- tate was centrifuged, washed twice with cold 1 per cent streptomycin, and dissolved in 0.40 ml. of 1.0 M Tris buffer, pH 8.0. The polymer solutions thus obtained were highly viscous. Reversal of Reaction-The formation of polyadenylate was readily re- versed by adding Pi to the incubation mixture. Table I11 shows the extent of polyadenylate (S-C14) breakdown in the presence of graded phos- phate concentrations. Phosphorolysis of Nucleic Acids-The extent and rate of phosphorolysis of different nucleic acids are shown in Table IV.4 Highly polymerized yeast RNA, prepared according to the procedure of Crestfield et al. (18),5 and plant virus RNA were phosphorolyzed readily. On the other hand, a dialyzed commercial sample of yeast sodium nucleate (Schwarz Labora- tories, Jnc.) mas decomposed at only about one-fourth thc rate and to a more limited extent. (18 per (wit). Duplicate assays of each experiment 4 These experiments were carried out together with Dr. 1,. A. Heppel in his labora- tory at the National Institutes of Health. 6 Kindly given to Dr. Heppel by Dr. F. W. Allen. 1086 SYNTHESIS OF POLYRIBONUCLEOTIDES Pi concentration Y x 104 0.64* (60 min.) 4.40 (60 " ) 19.0 (60 " ) 0.64* (0 min.) 38.0 (60 " ) by paper chromatography (isopropanol-water 1: 3, v/v, with NHa in the vapor phase (19)), showed an accumulation of nucleoside diphosphates which appeared to be approximately proportional to the amount of P132 incorporated into the acid-soluble nucleotides. The non-dialyzable limit polynucleotide obtained after exhaustive digestion of commercial yeast RNA with pancreatic ribonuclease was phosphorolyzed very little if at all. When phosphate was omitted from the reaction mixtures, no degrada- tion of RNA was detectable. Polyadenylate 6.P.m. 9er cent brmkdown 315 0 272 14 223 29 13 06 3 99 TABLEI III Polyadenylate Phosphorolysis The incubation mixture (0.25 ml.) contained 0.01 ml. of polyadenylate (8-04, containing 3.7 pmoles of adenine residues per ml. and 31,500 c.p.m. per ml.), 0.05 ml. of glycylglycine buffer (1 M, pH 7.4), Pi as indicated, 0.01 mi. of MgClz (0.1 M), and 0.05 unit of Ethanol 11. The mixturewas incubated at 37" for 60 minutes. The reac- tion was stopped and the acid-insoluble fraction was separated and its radioac- tivity determined. Polyadenylate was prepared from ADP (8-C14, 8.34 X 103 c.p.m. per pmole) as described in the text. * Pi concentration of the reaction mixture without added Pi. Exchange of Inorganic Phosphate with Nucleoside Diphosphates Specificity with Diflerent Nucleoside Diphosphates-P,32 exchanged with several nucleoside diphosphates, and this reaction was used for assaying the enzyme activity (Assay C). ADP, UDP, and CDP were found to react readily, both in the crude and in the purified enzyme fraction. How- ever, with GDP the rate of exchange with Pi32 in the crude extract was considerably slower than with the other diphosphates, while in the puri- fied fraction almost no activity relative to the other diphosphates was detected (Table V). Furthermore, when the extent of polymerization was measured with the purified enzyme, it was found that ADP, UDP, and CDP were polymerized to a considerable extent, but under these con- ditions no acid-insoluble polymer of GDP was observed. GTP plus G5P in the presence of purified yeast nucleoside monophosphate kinase gave the same low rate of exchange as did GDP alone. Thymidine diphosphate showed only a feeble activity; 0.06 ml. of Ethanol I incorporated 0.17 xmole of Pj into TDP as compared with 12 mpmoles into ADP. U. Z. Ll`lvl'AUElt AND A. KORNBERC 1087 Time of incuba- hrs. 24 3 3 3 3 3 3 5 5 3 3 3 3 3 3 3 3 3 3 Influence os Various Factors on Rate of Exchange "Activating" Fuctor-When the enzyme fractions obtained during the course of purification were assayed for their PI3' exchange rate with ADP, TABLE IV Phosphorolysis of Nucleic Acids The incubation mixture (0.25 ml.) contained 0.02 ml. of Tris buffer (0.5 M, pH 8.0), 0.01 ml. of MgC12 (0.1 M), 0.06 ml. of potassium phosphate buffer (0.2 M, pH 7.2), 0.01 to 0.06 ml. of Pi32 solution, and amounts of RNA and enzyme (Ethanol I) as indicated. The mixture was incubated at 37". The reaction was stopped and treated with Norit A as described in Assay C. The amounts of P132 added (expressed as 106 c.p.m.) were 2.1, 4.1, 4.5, 15, 15, and 17, in Experiments 1 through 6, respec- $1;; rated pmole 0.41 0.07 0.26 0.62 0.24 0.02 0.05 0.25 0.26 0.019 0.027 0.59 0.95 0.17 0.31 0.056 0.039 0.030 0.15 incorpo- -~ tively. Experi- ment No. Polymer Purified yeast RNA (1717 (( (6